ATPAttribute Entity

An ATP Attribute is a named piece of data about a project. This entity contains the names; the actual data is kept in a relationship between this entity and ATProject.

Relationships

• ATPAttribute IsSpecifiedFor ATProject (many-to-many)

ATPAttribute Fields

Name	Type	Notes
id	string	Unique identifier for this ATPAttribute.

ATPAttribute Indexes

Table	Name	Type	Fields	Notes
ATPAttribute	idx0	unique	id	Primary index for ATPAttribute.

ATProject Entity

An alignment/tree project is an organized group of alignments and trees for protein sequences, usually related to a single subsystem and its roles.

Relationships

1. ATProject HasValueOf ATPAttribute (many-to-many)
2. ATProject IsProjectFor Alignment (one-to-many)
3. ATProject UsesProteinsIn RepresentativeSet (many-to-many)
4. ATProject IsSelectedWith Role (many-to-many)

ATProject Fields

Name	Type	Notes
id	string	Unique identifier for this ATProject.
version	string	Version number of this project, updated when the project is changed.

ATProject Indexes

Table	Name	Type	Fields	Notes
ATProject	idx0	unique	id	Primary index for ATProject.

AliasType Entity

An alias type describes a kind of alternate name for a compound or reaction.

Relationships

1. AliasType IsAliasOf Compound (many-to-many)
2. AliasType IsAliasOfReaction Reaction (many-to-many)

AliasType Fields

Name	Type	Notes
id	string	Unique identifier for this AliasType.
source	string	This is the source associated with this alias type. It may be a pubmed ID or a website.

AliasType Indexes

Table	Name	Type	Fields	Notes
AliasType	idx0	unique	id	Primary index for AliasType.

Alignment Entity

An alignment is a set of protein sequences, each associated with an ID and a comment, that is organized so that the individual amino acids line up in a meaningful way.

Relationships

1. Alignment IsOrganizedBy AlignmentTree (one-to-many)
2. Alignment AlignsProteinsIn RepresentativeSet (many-to-one)
3. Alignment IsPartOf ATProject (many-to-one)

Alignment Fields

Name	Type	Notes
id	int	Unique identifier for this Alignment.
alignment	proteinData	Content of the alignment, essentially an ordered list of (ID, comment, sequence) triples.
description	text	Description of the parameters and tools used to generate this alignment.
trimmed	boolean	TRUE if this alignment is trimmed, in other words if it aligns subsets of the original sequences instead of the entire sequences.

Alignment Indexes

Table	Name	Type	Fields	Notes
Alignment	idx0	unique	id	Primary index for Alignment.

AlignmentTree Entity

A tree is an acyclic graph of the sequences in an alignment, organized according to a particular principle. A single alignment can have many trees, some of which are rooted and some of which are unrooted.

Relationships

• AlignmentTree Organizes Alignment (many-to-one)

AlignmentTree Fields

Name	Type	Notes
id	int	Unique identifier for this AlignmentTree.
description	text	Description of the parameters and tools used to generate this tree.
root	string	The name of the root node, or an empty string if the tree is unrooted.
tree-text	text	A string representation of the tree in NEWICK format.

AlignmentTree Indexes

Table	Name	Type	Fields	Notes
AlignmentTree	idx0	unique	id	Primary index for AlignmentTree.

Annotation Entity

An annotation is a comment attached to a feature. Annotations are used to track the history of a feature's functional assignments and any related issues. The key is the feature ID followed by a colon and an complemented ten-digit sequence number. The complemented sequence number causes the annotations to sort with the most recent one first.

Relationships

• Annotation Annotates Feature (many-to-one)

Annotation Fields

Name	Type	Notes
id	string	Unique identifier for this Annotation.
annotation-time	date	Date and time at which the annotation was made.
annotator	string	Name of the annotator who made the comment.
comment	text	Text of the annotation.

Annotation Indexes

Table	Name	Type	Fields	Notes
Annotation	idx0	unique	id	Primary index for Annotation.

AtomicRegulon Entity

An atomic regulon is an indivisible group of coregulated features on a single genome. Atomic regulons are constructed so that a given feature can only belong to one. Because of this, the expression levels for atomic regulons represent in some sense the state of a cell. The ID of an atomic regulon is the genome ID, a colon, and the atomic regulon sequence number.

Relationships

1. AtomicRegulon IsFormedOf Feature (one-to-many)
2. AtomicRegulon IsAffectedIn Experiment (many-to-many)
3. AtomicRegulon WasGeneratedFrom Chip (many-to-many)
4. AtomicRegulon ReflectsStateOf Genome (many-to-one)

AtomicRegulon Fields

Name	Type	Notes
id	string	Unique identifier for this AtomicRegulon.

AtomicRegulon Indexes

Table	Name	Type	Fields	Notes
AtomicRegulon	idx0	unique	id	Primary index for AtomicRegulon.

Attribute Entity

An attribute describes a category of condition or characteristic for an experiment. The goals of the experiment can be inferred from its values for all the attributes of interest.

Relationships

• Attribute HasValueIn Experiment (many-to-many)

Attribute Fields

Name	Type	Notes
id	string	Unique identifier for this Attribute.
description	text	Descriptive text indicating the nature and use of this attribute.

Attribute Indexes

Table	Name	Type	Fields	Notes
Attribute	idx0	unique	id	Primary index for Attribute.

Chip Entity

A chip is a device containing multiple probe sequences for use in gene expression experiments.

Relationships

1. Chip GeneratedLevelsFor AtomicRegulon (many-to-many)
2. Chip HasResultsIn Experiment (one-to-many)
3. Chip IndicatedLevelsFor Feature (many-to-many)
4. Chip ProducedResultsFor Genome (many-to-many)

Chip Fields

Name	Type	Notes
id	string	Unique identifier for this Chip.

Chip Indexes

Table	Name	Type	Fields	Notes
Chip	idx0	unique	id	Primary index for Chip.

Cluster Entity

A cluster is a set of features that occur close to each other on a genome and are related by a significant degree of functional coupling.

Relationships

• Cluster IsOccurrenceOf Feature (many-to-many)

Cluster Fields

Name	Type	Notes
id	int	Unique identifier for this Cluster.

Cluster Indexes

Table	Name	Type	Fields	Notes
Cluster	idx0	unique	id	Primary index for Cluster.

Compound Entity

A compound is a chemical that participates in a reaction. All compounds have a unique ID and may also have one or more names. Both ligands and reaction components are treated as compounds.

Relationships

1. Compound HasAliasOf AliasType (many-to-many)
2. Compound IsPresentIn Media (many-to-many)
3. Compound IsTerminusFor Scenario (many-to-many)
4. Compound IsAttractedTo Structure (many-to-many)
5. Compound IsInvolvedIn Reaction (many-to-many)
6. Compound IsComposedOf StructuralCue (many-to-many)
7. Compound IsShownOn Diagram (many-to-many)

Compound Fields

Name	Type	Notes
id	string	Unique identifier for this Compound.
charge	int	The charge of the compound in aqueous solution at pH7.
chargedFormula	string	The molecular formula of the charged form of the compound at pH7.
deltaG	float	The Gibbs free energy change of formation estimated using the group contribution method. The units are kcal/mol.
deltaGerr	float	The uncertainty in the Gibbs free energy change of formation estimated using the group contribution method. The units are kcal/mol.
hasMolfile	boolean	This boolean is true only if a molfile exists for the compound.
label	string	Primary name of the compound. This is the name used in reaction display strings.
mass	float	The molecular weight of the compound.
neutralFormula	string	The molecular formula of the compound in its neutral form.
stringcode	text	This string uniquely represents the 2D structure of a compound. The stringcode can be used to query the compound database to find compounds that match a particular 2D structures. Only stereoisomers will have identical stringcodes.

Compound Indexes

Table	Name	Type	Fields	Notes
Compound	idx0		label	This index allows searching for compounds by name.
Compound	idx1	unique	id	Primary index for Compound.

Contig Entity

A contig is a contiguous sequence of base pairs belonging to a single genome. The key of the contig is the genome ID followed by a colon and then the contig ID.

Relationships

1. Contig HasSection DNASequence (one-to-many)
2. Contig IsLocusFor Feature (many-to-many)
3. Contig MakesUp Genome (many-to-one)

Contig Fields

Name	Type	Notes
id	string	Unique identifier for this Contig.
length	counter	Number of base pairs in the contig.

Contig Indexes

Table	Name	Type	Fields	Notes
Contig	idx0	unique	id	Primary index for Contig.

CoregulatedSet Entity

A coregulated set is a group of features that are believed to express up or down in concert. The coregulation is inferred from the experimental expression' data.

Relationships

• CoregulatedSet Regulates Feature (many-to-many)

CoregulatedSet Fields

Name	Type	Notes
id	hash-string	Unique identifier for this CoregulatedSet.
reason	text	Description of how this coregulated set was derived.

CoregulatedSet Indexes

Table	Name	Type	Fields	Notes
CoregulatedSet	idx0	unique	id	Primary index for CoregulatedSet.

DNASequence Entity

A DNA sequence is a segment of a contig. Contigs are broken into multiple DNA sequences in order to prevent problems that might arise from loading an entire contig into memory when it consists of a billion or more base pairs. The maximum length of a DNA sequence is one million bases. The key is the contigID followed by a 7-digit ordinal number. So, the first sequence will have an ordinal of 0000000 and contain the base pairs from position 1 to position 1,000,000, the second will have an ordinal of 0000001 and contain the pairs from position 1,000,001 to 2,000,000, and so forth. This will allow us to store 10 trillion base pairs per contig.

Relationships

• DNASequence IsSectionOf Contig (many-to-one)

DNASequence Fields

Name	Type	Notes
id	string	Unique identifier for this DNASequence.
sequence	dna	Base pairs that make up this sequence.

DNASequence Indexes

Table	Name	Type	Fields	Notes
DNASequence	idx0	unique	id	Primary index for DNASequence.

Diagram Entity

A functional diagram describes a network of chemical reactions, often comprising a single subsystem. A diagram is identified by a short name and contains a longer descriptive name.

Relationships

1. Diagram Displays Reaction (many-to-many)
2. Diagram IsRelevantFor Subsystem (many-to-many)
3. Diagram Shows Compound (many-to-many)
4. Diagram IncludesPartOf Scenario (many-to-many)

Diagram Fields

Name	Type	Notes
id	string	Unique identifier for this Diagram.
name	text	Descriptive name of this diagram.
content	image array	The content of the diagram, in PNG format.

Diagram Indexes

Table	Name	Type	Fields	Notes
Diagram	idx0	unique	id	Primary index for Diagram.

EcNumber Entity

EC numbers are assigned by the Enzyme Commission, and consist of four numbers separated by periods, each indicating a successively smaller cateogry of enzymes.

Relationships

1. EcNumber IsConsistentWith Role (many-to-many)
2. EcNumber Categorizes Reaction (many-to-many)

EcNumber Fields

Name	Type	Notes
id	string	Unique identifier for this EcNumber.
obsolete	boolean	This boolean indicates when an EC number is obsolete.
replacedby	string	When an obsolete EC number is replaced with another EC number, this string will hold the name of the replacement EC number.

EcNumber Indexes

Table	Name	Type	Fields	Notes
EcNumber	idx0	unique	id	Primary index for EcNumber.

Experiment Entity

An experiment is a combination of conditions for which gene expression information is desired. The result of the experiment is a set of expression levels for features under the given conditions.

Relationships

1. Experiment AffectsLevelOf AtomicRegulon (many-to-many)
2. Experiment HasValueFor Attribute (many-to-many)
3. Experiment IndicatesSignalFor Feature (many-to-many)
4. Experiment OperatesIn Media (many-to-many)
5. Experiment HasResultsFor Chip (many-to-one)

Experiment Fields

Name	Type	Notes
id	string	Unique identifier for this Experiment.
source	string	Publication or lab relevant to this experiment.

Experiment Indexes

Table	Name	Type	Fields	Notes
Experiment	idx0	unique	id	Primary index for Experiment.

Family Entity

A family is a group of features united by a particular determination algorithm. The algorithm will frequently-- but not always-- signify a functional role. The key is a character code for the kind of family (generally two or three letters) followed by the family's ID number.

Relationships

1. Family HasMember Feature (many-to-many)
2. Family IsCoupledTo Family (many-to-many)
3. Family IsFamilyFor Role (many-to-many)
4. Family IsRepresentedIn Genome (many-to-many)
5. Family IsCoupledWith Family (many-to-many)

Family Fields

Name	Type	Notes
id	string	Unique identifier for this Family.
family-function	text array	Optional free-form description of the family. For function-based families, this would be the functional role for the family members. For other family types it could be a common name or an alias.

Family Indexes

Table	Name	Type	Fields	Notes
Family	idx0	unique	id	Primary index for Family.

Feature Entity

A feature (sometimes also called a gene) is a part of a genome that is of special interest. Features may be spread across multiple DNA sequences (contigs) of a genome, but never across more than one genome. Each feature in the database has a unique FIG ID that functions as its ID in this table.

Relationships

1. Feature IsAnnotatedBy Annotation (one-to-many)
2. Feature IsAttachmentSiteFor Feature (many-to-many)
3. Feature IsCoregulatedWith Feature (many-to-many)
4. Feature IsInPair Pairing (many-to-many)
5. Feature IsLocatedIn Contig (many-to-many)
6. Feature IsRegulatedWith CoregulatedSet (many-to-many)
7. Feature OccursIn Cluster (many-to-many)
8. Feature IsContainedIn MachineRole (many-to-many)
9. Feature IsMemberOf Family (many-to-many)
10. Feature HasProteinIn RepresentativeSet (many-to-many)
11. Feature IsIdentifiedBy Identifier (many-to-many)
12. Feature HasLevelsFrom Chip (many-to-many)
13. Feature HasIndicatedSignalFrom Experiment (many-to-many)
14. Feature HasAttachmentSite Feature (many-to-many)
15. Feature HasCoregulationWith Feature (many-to-many)
16. Feature IsFormedInto AtomicRegulon (many-to-one)
17. Feature HasFunctional Role (many-to-many)
18. Feature IsOwnedBy Genome (many-to-one)
19. Feature Produces ProteinSequence (many-to-one)
20. Feature IsInSet FeatureSet (many-to-many)

Feature Fields

Name	Type	Notes
id	string	Unique identifier for this Feature.
feature-type	string	Code indicating the type of this feature. Among the codes currently supported are "peg" for a protein encoding gene, "bs" for a binding site, "opr" for an operon, and so forth.
function	text	Functional assignment for this feature. This will often indicate the feature's functional role or roles, and may also have comments.
locked	boolean	If TRUE, then this feature is locked and its functional role assignment cannot be changed.
sequence-length	counter	Number of base pairs in this feature.
essential	link array	A value indicating the essentiality of the feature, In most cases, this will be a word describing whether the essentiality is confirmed (essential) or potential (potential-essential), hyperlinked to the document from which the essentiality was curated. If a feature is not essential, this field will have no values; otherwise, it may have multiple values.
evidence-code	string array	An evidence code describes the possible evidence that exists for deciding a feature's functional assignment. A feature may have no evidence, a single evidence code, or several.
link	text array	Web hyperlink for this feature. A feature can have no hyperlinks or it can have many. The links are to other websites that have useful about the gene that the feature represents, and are coded as raw HTML, using an anchor href tag.
virulent	link array	A value indicating the virulence of the feature, coded as HTML. In most cases, this will be a phrase or SA number hyperlinked to the document from which the virulence information was curated. If the feature is not virulent, this field will have no values; otherwise, it may have multiple values.

Feature Indexes

Table	Name	Type	Fields	Notes
Feature	idx0	unique	id	Primary index for Feature.
FeatureEvidence	idx0		evidence-code, id	This index is used to find features by evidence code.

FeatureSet Entity

A feature set is a group of features that implement a reaction in a particular model. The key is a hash string computed from the model name and a sorted list of the initial features in the set.

Relationships

1. FeatureSet IsSetOf Feature (many-to-many)
2. FeatureSet IsEncompassedIn Model (many-to-one)
3. FeatureSet Stimulates Reaction (many-to-many)

FeatureSet Fields

Name	Type	Notes
id	hash-string	Unique identifier for this FeatureSet.
source	string	This is the source of the feature set.

FeatureSet Indexes

Table	Name	Type	Fields	Notes
FeatureSet	idx0	unique	id	Primary index for FeatureSet.

Genome Entity

A genome represents a specific organism with DNA, or a specific meta-genome. All DNA sequences in the database belong to genomes.

Relationships

1. Genome HasRepresentativeOf Family (many-to-many)
2. Genome IsConfiguredBy AtomicRegulon (one-to-many)
3. Genome IsMadeUpOf Contig (one-to-many)
4. Genome IsModeledBy Model (one-to-many)
5. Genome IsOwnerOf Feature (one-to-many)
6. Genome Uses MolecularMachine (one-to-many)
7. Genome IsCollectedInto GenomeSet (many-to-one)
8. Genome IsInTaxa TaxonomicGrouping (many-to-one)
9. Genome HadResultsProducedBy Chip (many-to-many)

Genome Fields

Name	Type	Notes
id	string	Unique identifier for this Genome.
complete	boolean	TRUE if the genome is complete, else FALSE
contigs	int	Number of contigs for this genome.
dna-size	counter	Number of base pairs in the genome.
domain	string	Domain for this organism (Archaea, Bacteria, Eukaryota, Virus, Plasmid, or Environmental Sample).
genetic-code	int	Genetic code number used for protein translation on most of this genome's contigs.
pegs	int	Number of protein encoding genes for this genome.
prokaryotic	boolean	TRUE if this is a prokaryotic genome, else FALSE
rnas	int	Number of RNA features found for this organism.
scientific-name	string	Full genus/species/strain name of the genome.

Genome Indexes

Table	Name	Type	Fields	Notes
Genome	idx0		scientific-name	This index allows the applications to find all genomes in lexical order by name.
Genome	idx1	unique	id	Primary index for Genome.

GenomeSet Entity

A genome set is a named group of related genomes. Each genome set consists of genomes that use highly similar ribosomal small subunits. Two genomes are in the same set if they have a similarity of 97% or greater in subunits of length 1000 or more.

Relationships

• GenomeSet IsCollectionOf Genome (one-to-many)

GenomeSet Fields

Name	Type	Notes
id	string	Unique identifier for this GenomeSet.

GenomeSet Indexes

Table	Name	Type	Fields	Notes
GenomeSet	idx0	unique	id	Primary index for GenomeSet.

Identifier Entity

An identifier is an alternate name for a feature or a protein sequence. Identifiers are preferentially associated with features; however, in some cases the precise feature named by an external identifier cannot be computed; in this case, the identifier is associated with the protein sequence. Identifiers are stored in a prefixed format that insures identifiers from different sources have different IDs.

Relationships

1. Identifier HasAssertionFrom Source (many-to-many)
2. Identifier Identifies Feature (many-to-many)
3. Identifier Names ProteinSequence (many-to-many)

Identifier Fields

Name	Type	Notes
id	string	Unique identifier for this Identifier.
natural-form	string	Natural form of the identifier. This is how the identifier looks without the identifying prefix (if one is present).
source	string	Specific type of the identifier, such as its source database or category. The type can usually be decoded to convert the identifier to a URL.

Identifier Indexes

Table	Name	Type	Fields	Notes
Identifier	idx0		source, natural-form	This index allows all the identifiers of a specified type to be located.
Identifier	idx1		natural-form	This index allows looking up a natural identifier (that is, one without the identifying prefix..
Identifier	idx2	unique	id	Primary index for Identifier.

MachineRole Entity

A machine role represents a role as it occurs in a molecular machine. The key is a colon-delimited triple containing an MD5 hash of the subsystem ID followed by a genome ID (with optional region string) and a role abbreviation. The machine role corresponds to a cell on the subsystem spreadsheet. Features in the subsystem are assigned directly to the machine role.

Relationships

1. MachineRole Contains Feature (many-to-many)
2. MachineRole IsRoleFor MolecularMachine (many-to-one)
3. MachineRole HasRole Role (many-to-one)

MachineRole Fields

Name	Type	Notes
id	string	Unique identifier for this MachineRole.

MachineRole Indexes

Table	Name	Type	Fields	Notes
MachineRole	idx0	unique	id	Primary index for MachineRole.

Media Entity

A media describes the chemical content of the solution in which cells are grown in an experiment or for the purposes of a model. The key is the common media name. The nature of the media is described by its relationship to its constituent compounds.

Relationships

1. Media IsComprisedOf Compound (many-to-many)
2. Media IsUtilizedIn Experiment (many-to-many)

Media Fields

Name	Type	Notes
id	string	Unique identifier for this Media.
defined	boolean	A media is considered defined if the exact chemical content is known.
source	string	Publication or laboratory that provided the source information for the media.

Media Indexes

Table	Name	Type	Fields	Notes
Media	idx0	unique	id	Primary index for Media.

Model Entity

A model specifies a relationship between sets of features and reactions in a cell. It is used to simulate cell growth and gene knockouts to validate annotations.

Relationships

1. Model Encompasses FeatureSet (one-to-many)
2. Model Models Genome (many-to-one)
3. Model Requires Reaction (many-to-many)

Model Fields

Name	Type	Notes
id	string	Unique identifier for this Model.
buildDate	date	The date when the model was built.
class	string	This indicates the type of the organism being modeled: gram negative or gram positive for now.
directory	string	The computer directory where the model data is stored.
gapFillDate	date	The date when the gap filling was performed.
gapFillingMinimalObjective	float	The minimal objective projected for the gap filling problem.
gapFillingObjective	float	The objective reached with the final solution of the gap filling problem.
gapFillingReactions	int	The number of gap filling reactions.
gapFillingTime	float	The time required to arrive at the gap filling solution.
gapFillVersion	int	The version number of the gapfilling of the model.
genesWithFunction	int	This indicates the number of genes with known functions in the organism being modeled.
genesWithReactions	int	This indicates the number of genes with reactions associated with them in the model.
gramNegativeGenes	int	This indicates the total number of gram negative genes in the organism being modeled.
gramPositiveGenes	int	This indicates the total number of gram positive genes in the organism being modeled.
jobid	int	For models of RAST genomes, this is the job id associated with the RAST genome.
message	string	Short message describing the status of the model.
modificationTime	date	Date and time of the last modification to the model.
nonsubsystemGenes	int	This indicates the number of genes not in a subsystem.
nonsubsystemGenesWithReactions	int	This indicates the number of genes not in a subsystem that have reactions associated with them.
numberOfReactions	int	The number of genes that have reactions associated with them.
numberOfTransporters	int	The number of transport reactions in the model.
obsolete	boolean	This boolean equals true if the model is obsolete.
organismName	string	The name of the organism associated with this model.
owner	string	This is user ID or name of the user who owns this model.
source	string	This indicates the source of the model. For published models, this will be a pubmed ID. For automatically generated SEED models, it will be 'SEED'.
status	int	Code indicating the status of the model (e.g. -1 is being built, 0 is gap-filling, etc.).
subsystemGenes	int	This indicates the number of genes that are assigned to a subsystem in the organism being modeled.
subsystemGenesWithReactions	int	This indicates the number of genes in subsystems that are associated with a reaction.
totalGenes	int	This indicates the total number of genes in the organism being modeled.
version	int	The version number of the model.
users	string array	ID or name of a user who has access to this model.

Model Indexes

Table	Name	Type	Fields	Notes
Model	idx0	unique	id	Primary index for Model.

MolecularMachine Entity

A molecular machine is a collection of features that implements a metabolic pathway. Machines are the physical instances of variants. Each machine corresponds to a row in a subsystem spreadsheet. The key is an MD5 hash formed from a colon-separated list containing the subsystem ID, the variant ID, the Genome ID, and an optional region string. It is possible for a single subsystem to occur multiple times in a particular genome. If that is the case, then the subsystem will have a molecular machine for each occurrence. The region string used in forming the key insures that each machine has a unique ID.

Relationships

1. MolecularMachine IsMachineOf MachineRole (one-to-many)
2. MolecularMachine Implements Variant (many-to-one)
3. MolecularMachine IsUsedBy Genome (many-to-one)

MolecularMachine Fields

Name	Type	Notes
id	hash-string	Unique identifier for this MolecularMachine.
curated	boolean	This flag is TRUE if the assignment of the molecular machine has been curated, and FALSE if it was made by an automated program.
region	string	Region in the genome for which the machine is relevant. Normally, this is an empty string, indicating that the machine covers the whole genome. If a subsystem has multiple machines for a genome, this contains a location string describing the region occupied by this particular machine.

MolecularMachine Indexes

Table	Name	Type	Fields	Notes
MolecularMachine	idx0	unique	id	Primary index for MolecularMachine.

PairSet Entity

A pair set indicates evidence for a functional connection between protein sequence pairs. The protein sequences possessing the connection are the ones that participate in the evidence set's pairings. The pairings for a particular evidence set will contain protein sequences that are significantly similar. In other words, if (A,B) and (X,Y) are both pairings in a single evidence set, then (A =~ X) and (B =~ Y) or (A =~ Y) and (B =~ X), depending on the value of the "inverted" attribute of the IsDeterminedBy relationship. Essentially, a pairing in its own right is unordered. If (A,B) is a pair, then so is (B,A). However, the evidence set maintains a correspondence between its pairs that _is_ ordered, because the constituent pairs must match. The direction in which a pair matches others in the set is an attribute of the relationship from the pairs to the sets.

Relationships

• PairSet IsDeterminedBy Pairing (one-to-many)

PairSet Fields

Name	Type	Notes
id	int	Unique identifier for this PairSet.
score	int	Score for this evidence set. The score indicates the number of significantly different genomes represented by the pairings.

PairSet Indexes

Table	Name	Type	Fields	Notes
PairSet	idx0	unique	id	Primary index for PairSet.

Pairing Entity

A pairing indicates that two features are found close together in a genome. Not all possible pairings are stored in the database; only those that are considered for some reason to be significant for annotation purposes.The key of the pairing is the concatenation of the feature IDs in alphabetical order with an intervening colon. Theoretically, the pairing is unordered: (A,B) and (B,A) are the same pairing. It is frequently the case, however, that we need to refer to the "first" or "second" protein in the pairing. When this happens, the first one is always the protein with the alphabetically lesser key. The IsInPair relationship automatically shows the proteins in this order.

Relationships

1. Pairing Determines PairSet (many-to-one)
2. Pairing IsPairOf Feature (many-to-many)

Pairing Fields

Name	Type	Notes
id	string	Unique identifier for this Pairing.

Pairing Indexes

Table	Name	Type	Fields	Notes
Pairing	idx0	unique	id	Primary index for Pairing.

ProteinSequence Entity

A protein sequence is a specific sequence of amino acids. Unlike a DNA sequence, a protein sequence does not belong to a genome. Identical proteins generated by different genomes are generally stored as a single ProteinSequence instance. The key is a hash code computed from the protein letter sequence.

Relationships

1. ProteinSequence Exposes Structure (many-to-many)
2. ProteinSequence IsNamedBy Identifier (many-to-many)
3. ProteinSequence IsProteinFor Feature (one-to-many)
4. ProteinSequence IsATopicOf Publication (many-to-many)

ProteinSequence Fields

Name	Type	Notes
id	string	Unique identifier for this ProteinSequence.
sequence	text	The sequence contains the letters corresponding to the protein's amino acids.

ProteinSequence Indexes

Table	Name	Type	Fields	Notes
ProteinSequence	idx0	unique	id	Primary index for ProteinSequence.

Publication Entity

A publication is an article or citation that may be used as evidence for assertions made in the database. The key is the PUBMED ID.

Relationships

• Publication Concerns ProteinSequence (many-to-many)

Publication Fields

Name	Type	Notes
id	string	Unique identifier for this Publication.
citation	link	Hyperlink of the article. The text is the article title.

Publication Indexes

Table	Name	Type	Fields	Notes
Publication	idx0		citation	This index provides the ability to search by article title. It should only be used for LIKE-style searches, since the article titles are encoded together with the URLs.
Publication	idx1	unique	id	Primary index for Publication.

Reaction Entity

A reaction is a chemical process that converts one set of compounds (substrate) to another set (products). The reaction ID is generally a small number preceded by a letter.

Relationships

1. Reaction HasReactionAliasOf AliasType (many-to-many)
2. Reaction Involves Compound (many-to-many)
3. Reaction IsCategorizedInto EcNumber (many-to-many)
4. Reaction IsRequiredBy Model (many-to-many)
5. Reaction IsStimulatedBy FeatureSet (many-to-many)
6. Reaction IsTriggeredBy RoleSet (many-to-many)
7. Reaction IsDisplayedOn Diagram (many-to-many)
8. Reaction ParticipatesIn Scenario (many-to-many)

Reaction Fields

Name	Type	Notes
id	string	Unique identifier for this Reaction.
code	text	This is a unique string representing all reaction data. This string is used to determine if two reactions are matching.
definition	text	This string contains the entire reaction equation written out for display.
deltaG	float	The Gibbs free energy change of reaction estimated using the group contribution method. The units are kcal/mol.
deltaGerr	float	The unicertainty in the Gibbs free energy change of reaction estimated using the group contribution method. The units are kcal/mol.
keggMap	text	A ; delimited string listing the KEGG maps that include this reaction.
MMdeltaG	float	The millimolar Gibbs free energy change of reaction estimated using the group contribution method. The units are kcal/mol.
operator	text	This string describes the reaction rules that are capable of reproducing the reaction.
pathway	text	A ; delimited string listing the pathways this reaction takes part in.
predictedReversibility	char	This is the reversibility predicted for the reaction based on the estimated deltaG and a set of heuristic rules. There are three possible values: B for backward, F for forward, and R for reversible.

Reaction Indexes

Table	Name	Type	Fields	Notes
Reaction	idx0	unique	id	Primary index for Reaction.

RepresentativeSet Entity

A Representative Set is a set of proteins (usually also features) that are organized by sequence similarity or role. Each alignment in a project will relate to one representative set.

Relationships

1. RepresentativeSet ContainsProteinsFor Alignment (one-to-many)
2. RepresentativeSet HasProteinFor Feature (many-to-many)
3. RepresentativeSet ContainsProteinsUsedBy ATProject (many-to-many)

RepresentativeSet Fields

Name	Type	Notes
id	hash-string	Unique identifier for this RepresentativeSet.
description	text	Description of what this representative set means.

RepresentativeSet Indexes

Table	Name	Type	Fields	Notes
RepresentativeSet	idx0	unique	id	Primary index for RepresentativeSet.

Role Entity

A role describes a biological function that may be fulfilled by a feature. One of the main goals of the database is to assign features to roles. Most roles are effected by the construction of proteins. Some, however, deal with functional regulation and message transmission. A role represents a single gene function. Only roles in subsystems are represented in this way. The key of the role is a hash code computed from the name.

Relationships

1. Role IsFunctionalIn Feature (many-to-many)
2. Role IsRoleOf MachineRole (one-to-many)
3. Role IsSelectedBy ATProject (many-to-many)
4. Role IsIncludedIn Subsystem (many-to-many)
5. Role IsInCombination RoleSet (many-to-many)
6. Role IsConsistentTo EcNumber (many-to-many)
7. Role DeterminesFunctionOf Family (many-to-many)

Role Fields

Name	Type	Notes
id	string	Unique identifier for this Role.
hypothetical	boolean	TRUE if a role is hypothetical, else FALSE
role-index	int array	index of this role in role vectors, or -1 if the role does not appear in role vectors

Role Indexes

Table	Name	Type	Fields	Notes
Role	idx0	unique	id	Primary index for Role.
RoleIndex	idx0		role-index	This index is used to find roles by role-index.

RoleSet Entity

A role set is a group of roles that work together to stimulate a reaction. Most role sets consist of a single role; however, some reactions require the presence of multiple roles to get them started. A reaction is usually triggered by a single role, but some reactions are triggered by a boolean combination of roles (e.g. =(A and (B or C) and D) or (E and B and F) or G=). The boolean expression can be converted into disjunctive normal form, which is a list of alternative sets (e.g. =(A and B and D) or (A and C and D) or (E and B and F) or G=). Each alternative is then converted into a role set. This allows us to precisely represent the triggering conditions of a reaction in the database.

Relationships

1. RoleSet IsCombinationOf Role (many-to-many)
2. RoleSet Triggers Reaction (many-to-many)

RoleSet Fields

Name	Type	Notes
id	hash-string	Unique identifier for this RoleSet.
source	string	The source of the role set.

RoleSet Indexes

Table	Name	Type	Fields	Notes
RoleSet	idx0	unique	id	Primary index for RoleSet.

Scenario Entity

A scenario is a partial instance of a subsystem with a defined set of reactions. Each scenario converts input compounds to output compounds using reactions. The scenario may use all of the reactions controlled by a subsystem or only some, and may also incorporate additional reactions. Because scenario names are not unique, the actual scenario ID is a number.

Relationships

1. Scenario HasParticipant Reaction (many-to-many)
2. Scenario Overlaps Diagram (many-to-many)
3. Scenario Validates Subsystem (many-to-one)
4. Scenario HasAsTerminus Compound (many-to-many)

Scenario Fields

Name	Type	Notes
id	int	Unique identifier for this Scenario.
common-name	string	Common name of the scenario. The name, rather than the ID number, is usually displayed everywhere.

Scenario Indexes

Table	Name	Type	Fields	Notes
Scenario	idx0		common-name	This index allows the user to search for a scenario by name.
Scenario	idx1	unique	id	Primary index for Scenario.

Source Entity

A source is a user or organization that is permitted to make assertions about identifiers.

Relationships

• Source Asserts Identifier (many-to-many)

Source Fields

Name	Type	Notes
id	string	Unique identifier for this Source.

Source Indexes

Table	Name	Type	Fields	Notes
Source	idx0	unique	id	Primary index for Source.

StructuralCue Entity

A structural cue is a molecular substructure for a compound that is useful to known about.

Relationships

• StructuralCue IsFoundIn Compound (many-to-many)

StructuralCue Fields

Name	Type	Notes
id	string	Unique identifier for this StructuralCue.
charge	int	This is the charge of the substructure represented by this structural cue.
deltaG	float	This is the Gibbs free energy value associated with the structural cue in the group contribution method. This energy is used to estimate the deltaG of compounds involving the structural cue. The units are kcal/mol.
deltaGerr	float	This is the uncertainty in the free energy value associated with this structural cue in the group contribution method. This value is used to determine the uncertainty in the estimated delta values for compounds and reactions. The units are kcal/mol.
formula	string	This is the formula of the substructure represented by this structural cue.
priority	int	This is the priority with which the atom labeling algorithm searches for this structural cue in a molecular structure.
smallMolecule	boolean	Some structural cues represent entire molecules instead of structures. This feild will equal true when this is the case.
subgroups	string	Some structural cues can be further broken down into smaller structural cues. This field is a ; delimited list of the cues that this cue can be broken down into.

StructuralCue Indexes

Table	Name	Type	Fields	Notes
StructuralCue	idx0	unique	id	Primary index for StructuralCue.

Structure Entity

A structure is the geometrical representation of a protein sequence. A single protein sequence may have multiple structural representations, either because it is folded in different ways or because there are alternative representation formats. The key field is the representation type (currently only PDB) followed by a colon and the ID.

Relationships

1. Structure Attracts Compound (many-to-many)
2. Structure IsExposedBy ProteinSequence (many-to-many)

Structure Fields

Name	Type	Notes
id	string	Unique identifier for this Structure.

Structure Indexes

Table	Name	Type	Fields	Notes
Structure	idx0	unique	id	Primary index for Structure.

Subsystem Entity

A subsystem is a collection of roles that work together in a cell. Identification of subsystems is an important tool for recognizing parallel genetic features in different organisms. The key is subsystem name. The subsystem name used to come in two forms-- a natural form with spaces and an internal form with underscores. In this database we will only have the natural form.

Relationships

1. Subsystem Describes Variant (one-to-many)
2. Subsystem Includes Role (many-to-many)
3. Subsystem IsSubInstanceOf Scenario (one-to-many)
4. Subsystem IsInClass SubsystemClass (many-to-one)
5. Subsystem IsRelevantTo Diagram (many-to-many)

Subsystem Fields

Name	Type	Notes
id	string	Unique identifier for this Subsystem.
cluster-based	boolean	TRUE if this is a clustering-based subsystem, else FALSE. A clustering-based subsystem is one in which there is functional-coupling evidence that genes belong together, but we do not yet know what they do.
curator	string	Name of the person currently in charge of the subsystem.
description	text	Description of the subsystem's function in the cell.
experimental	boolean	TRUE if this is an experimental subsystem. An experimental subsystem is designed for investigation and is not yet ready to be used in comparative analysis and annotation.
notes	text	Descriptive notes about the subsystem.
private	boolean	TRUE if this is a private subsystem, else FALSE. A private subsystem has valid data, but is not considered ready for general distribution.
usable	boolean	TRUE if this is a usable subsystem, else FALSE. An unusable subsystem is one that is experimental or is of such low quality that it can negatively affect analysis.
version	int	Version number for the subsystem. This value is incremented each time the subsystem is backed up.

Subsystem Indexes

Table	Name	Type	Fields	Notes
Subsystem	idx0	unique	id	Primary index for Subsystem.

SubsystemClass Entity

Subsystem classes impose a hierarchical organization on the subsystems.

Relationships

1. SubsystemClass IsClassFor Subsystem (one-to-many)
2. SubsystemClass IsSuperclassOf SubsystemClass (one-to-many)
3. SubsystemClass IsSubclassOf SubsystemClass (many-to-one)

SubsystemClass Fields

Name	Type	Notes
id	string	Unique identifier for this SubsystemClass.

SubsystemClass Indexes

Table	Name	Type	Fields	Notes
SubsystemClass	idx0	unique	id	Primary index for SubsystemClass.

TaxonomicGrouping Entity

A taxonomic grouping is a segment of the classification for an organism. Taxonomic groupings are organized into a strict hierarchy by the IsGroupContaining relationship.

Relationships

1. TaxonomicGrouping IsGroupFor TaxonomicGrouping (one-to-many)
2. TaxonomicGrouping IsTaxonomyOf Genome (one-to-many)
3. TaxonomicGrouping IsInGroup TaxonomicGrouping (many-to-one)

TaxonomicGrouping Fields

Name	Type	Notes
id	counter	Unique identifier for this TaxonomicGrouping.
domain	boolean	TRUE if this is a domain grouping, else FALSE.
hidden	boolean	TRUE if this is a hidden grouping, else FALSE. Hidden groupings are not typically shown in a lineage list.
scientific-name	string	Primary scientific name for this grouping. This is the name used when displaying a taxonomy.
alias	string array	Alternate name for this grouping. A grouping may have many alternate names. The scientific name should also be in this list.

TaxonomicGrouping Indexes

Table	Name	Type	Fields	Notes
TaxonomicGrouping	idx0	unique	id	Primary index for TaxonomicGrouping.
TaxonomicGroupingAlias	idx0		alias	This index allows the user to find a particular taxonomic grouping by name. Because the scientifc name is also an alias, there is no index on scientific name.

Variant Entity

A variant is a functional subset of a subsystem. It indicates the particular sequence of roles used to implement a metabolic pathway. Variants are abstract concepts used to classify machines. The key of the variant is an MD5 hash of the subsystem ID followed by the variant code. The variant code is generally a number with zero or more decimal points, similar to what it done with software version numbers or legal outline numbers.

Relationships

1. Variant IsImplementedBy MolecularMachine (one-to-many)
2. Variant IsDescribedBy Subsystem (many-to-one)

Variant Fields

Name	Type	Notes
id	hash-string	Unique identifier for this Variant.
code	string	This is the variant code all by itself.
comment	text	Commentary text about the variant.
type	string	The variant type indicates the quality of the subsystem support. A type of "vacant" means that the subsystem does not appear to be implemented by the variant. A type of "incomplete" means that the subsystem appears to be missing many reactions. In all other cases, the type is "normal".
role-rule	text array	A space-delimited list of role IDs, in alphabetical order, that represents a possible list of non-auxiliary roles applicable to this variant. The roles are identified by their abbreviations. A variant may have multiple role rules.

Variant Indexes

Table	Name	Type	Fields	Notes
Variant	idx0	unique	id	Primary index for Variant.

AffectsLevelOf Relationship

This relationship indicates the expression level of an atomic regulon for a given experiment.

• Experiment AffectsLevelOf AtomicRegulon (many-to-many)
• AtomicRegulon IsAffectedIn Experiment (many-to-many)

AffectsLevelOf Fields

Name	Type	Notes
from-link	string	id of the source Experiment.
level	int	Indication of whether the feature is expressed (1), not expressed (-1), or unknown (0).
to-link	string	id of the target AtomicRegulon.

AffectsLevelOf Indexes

Table	Name	Type	Fields	Notes
AffectsLevelOf	idxFrom		from-link
AffectsLevelOf	idxTo		to-link

Attracts Relationship

This relationship connects a compound to the protein structures that attract it. This is an incomplete relationship that exists to service drug targeting queries. Only the attractions whose parameters have been determined through modeling or experimentation are included. The goal is to determine the docking energy between the compound and the protein structure.

• Structure Attracts Compound (many-to-many)
• Compound IsAttractedTo Structure (many-to-many)

Attracts Fields

Name	Type	Notes
electrostatic-energy	float	Docking energy in kcal/mol that results from the movement of electrons (electrostatic force) between the structure and the compound.
from-link	string	id of the source Structure.
reason	string	Indication of the reason for determining the docking energy. A value of "Random" indicates the docking was attempted as a part of a random survey used to determine the docking characteristics of a protein structure. A value of "Rich" indicates the docking was attempted because a low-energy docking result was predicted for the compound.
to-link	string	id of the target Compound.
tool	string	Name of the tool used to compute the docking energy.
total-energy	float	Total energy required for the compound to dock with the structure, in kcal/mol. A negative value means energy is released.
vanderwaals-energy	float	Docking energy in kcal/mol that results from the geometric fit (Van der Waals force) between the structure and the compound.

Attracts Indexes

Table	Name	Type	Fields	Notes
Attracts	idxFrom		from-link, total-energy	This index enables the application to view a structure's docking results from the lowest energy (best docking) to highest energy (worst docking).
Attracts	idxTo		to-link, total-energy	This index enables the application to view a compound's docking results from the lowest energy (best docking) to highest energy (worst docking).

Concerns Relationship

This relationship connects a publication to the protein sequences it describes.

• Publication Concerns ProteinSequence (many-to-many)
• ProteinSequence IsATopicOf Publication (many-to-many)

Concerns Fields

Name	Type	Notes
from-link	string	id of the source Publication.
to-link	string	id of the target ProteinSequence.

Concerns Indexes

Table	Name	Type	Fields	Notes
Concerns	idxFrom		from-link
Concerns	idxTo		to-link

Contains Relationship

This relationship connects a machine role to the features that occur in it. A feature may occur in many machine roles and a machine role may contain many features. The subsystem annotation process is essentially the maintenance of this relationship.

• MachineRole Contains Feature (many-to-many)
• Feature IsContainedIn MachineRole (many-to-many)

Contains Fields

Name	Type	Notes
from-link	string	id of the source MachineRole.
to-link	string	id of the target Feature.

Contains Indexes

Table	Name	Type	Fields	Notes
Contains	idxFrom		from-link
Contains	idxTo		to-link

ContainsProteinsFor Relationship

This relationship connects an alignment to the representative set containing the proteins aligned. There is one such set for each alignment.

• RepresentativeSet ContainsProteinsFor Alignment (one-to-many)
• Alignment AlignsProteinsIn RepresentativeSet (many-to-one)

ContainsProteinsFor Fields

Name	Type	Notes
from-link	hash-string	id of the source RepresentativeSet.
to-link	int	id of the target Alignment.

ContainsProteinsFor Indexes

Table	Name	Type	Fields	Notes
ContainsProteinsFor	idxFrom		from-link
ContainsProteinsFor	idxTo	unique	to-link

Describes Relationship

This relationship connects a subsystem to the individual variants used to implement it. Each variant contains a slightly different subset of the roles in the parent subsystem.

• Subsystem Describes Variant (one-to-many)
• Variant IsDescribedBy Subsystem (many-to-one)

Describes Fields

Name	Type	Notes
from-link	string	id of the source Subsystem.
to-link	hash-string	id of the target Variant.

Describes Indexes

Table	Name	Type	Fields	Notes
Describes	idxFrom		from-link
Describes	idxTo	unique	to-link

Displays Relationship

This relationship connects a diagram to its reactions. A diagram shows multiple reactions, and a reaction can be on many diagrams.

• Diagram Displays Reaction (many-to-many)
• Reaction IsDisplayedOn Diagram (many-to-many)

Displays Fields

Name	Type	Notes
from-link	string	id of the source Diagram.
location	rectangle	Location of the reaction's node on the diagram.
to-link	string	id of the target Reaction.

Displays Indexes

Table	Name	Type	Fields	Notes
Displays	idxFrom		from-link
Displays	idxTo		to-link

Encompasses Relationship

This relationship connects a model to its feature sets. A feature set can only be in a single model. The model can be said to be identified by its feature sets.

• Model Encompasses FeatureSet (one-to-many)
• FeatureSet IsEncompassedIn Model (many-to-one)

Encompasses Fields

Name	Type	Notes
from-link	string	id of the source Model.
to-link	hash-string	id of the target FeatureSet.

Encompasses Indexes

Table	Name	Type	Fields	Notes
Encompasses	idxFrom		from-link
Encompasses	idxTo	unique	to-link

Exposes Relationship

This relationship connects a protein sequence to its structural representations. It is a many-to-many relationship. Note that only some protein sequences have known structural representations.

• ProteinSequence Exposes Structure (many-to-many)
• Structure IsExposedBy ProteinSequence (many-to-many)

Exposes Fields

Name	Type	Notes
from-link	string	id of the source ProteinSequence.
to-link	string	id of the target Structure.

Exposes Indexes

Table	Name	Type	Fields	Notes
Exposes	idxFrom		from-link
Exposes	idxTo		to-link

GeneratedLevelsFor Relationship

This relationship connects an atomic regulon to a chip from which experimental data was produced for its features. It contains a vector of the expression levels.

• Chip GeneratedLevelsFor AtomicRegulon (many-to-many)
• AtomicRegulon WasGeneratedFrom Chip (many-to-many)

GeneratedLevelsFor Fields

Name	Type	Notes
from-link	string	id of the source Chip.
level-vector	countVector	Vector of expression levels (-1, 0, 1) for the experiments, in sequence order.
to-link	string	id of the target AtomicRegulon.

GeneratedLevelsFor Indexes

Table	Name	Type	Fields	Notes
GeneratedLevelsFor	idxFrom		from-link
GeneratedLevelsFor	idxTo		to-link

HasAliasOf Relationship

This relationship is used to find all the alternate names for a compound.

• Compound HasAliasOf AliasType (many-to-many)
• AliasType IsAliasOf Compound (many-to-many)

HasAliasOf Fields

Name	Type	Notes
alias	string	This is the actual alias used, and is frequently the main point of access for this relationship.
from-link	string	id of the source Compound.
to-link	string	id of the target AliasType.

HasAliasOf Indexes

Table	Name	Type	Fields	Notes
HasAliasOf	idx0		alias	This index can be used to find a particular alias when the type is not known.
HasAliasOf	idxFrom		from-link, alias	This index allows the application to find a particular alias if the type is known.
HasAliasOf	idxTo		to-link

HasAssertionFrom Relationship

Sources (users) can make assertions about identifiers using the annotation clearinghouse. When a user makes a new assertion about an identifier, it erases the old one.

• Identifier HasAssertionFrom Source (many-to-many)
• Source Asserts Identifier (many-to-many)

HasAssertionFrom Fields

Name	Type	Notes
expert	boolean	TRUE if this is an expert assertion, else FALSE
from-link	string	id of the source Identifier.
function	text	The function is the text of the assertion made about the identifier.
to-link	string	id of the target Source.

HasAssertionFrom Indexes

Table	Name	Type	Fields	Notes
HasAssertionFrom	idxFrom		from-link
HasAssertionFrom	idxTo		to-link

HasMember Relationship

This relationship connects each feature family to its constituent features. A family always has many features, and a single feature can be found in many families.

• Family HasMember Feature (many-to-many)
• Feature IsMemberOf Family (many-to-many)

HasMember Fields

Name	Type	Notes
from-link	string	id of the source Family.
to-link	string	id of the target Feature.

HasMember Indexes

Table	Name	Type	Fields	Notes
HasMember	idxFrom		from-link
HasMember	idxTo		to-link

HasParticipant Relationship

A scenario consists of many participant reactions that convert the input compounds to output compounds. A single reaction may participate in many scenarios.

• Scenario HasParticipant Reaction (many-to-many)
• Reaction ParticipatesIn Scenario (many-to-many)

HasParticipant Fields

Name	Type	Notes
from-link	int	id of the source Scenario.
to-link	string	id of the target Reaction.
type	int	Indicates the type of participaton. If 0, the reaction is in the main pathway of the scenario. If 1, the reaction is necessary to make the model work but is not in the subsystem. If 2, the reaction is part of the subsystem but should not be included in the modelling process.

HasParticipant Indexes

Table	Name	Type	Fields	Notes
HasParticipant	idxFrom		from-link, type	This index presents the reactions in the scenario in order from most important to least important.
HasParticipant	idxTo		to-link

HasProteinFor Relationship

This relationship connects a representative set to the features for its proteins. Note that there may be some proteins in the set which do not correspond to features. These are not represented.

• RepresentativeSet HasProteinFor Feature (many-to-many)
• Feature HasProteinIn RepresentativeSet (many-to-many)

HasProteinFor Fields

Name	Type	Notes
from-link	hash-string	id of the source RepresentativeSet.
to-link	string	id of the target Feature.

HasProteinFor Indexes

Table	Name	Type	Fields	Notes
HasProteinFor	idxFrom		from-link
HasProteinFor	idxTo		to-link

HasReactionAliasOf Relationship

This relationship is used to find all the alternate names for a reaction.

• Reaction HasReactionAliasOf AliasType (many-to-many)
• AliasType IsAliasOfReaction Reaction (many-to-many)

HasReactionAliasOf Fields

Name	Type	Notes
alias	string	This is the actual alias used, and is frequently the main point of access for this relationship.
from-link	string	id of the source Reaction.
to-link	string	id of the target AliasType.

HasReactionAliasOf Indexes

Table	Name	Type	Fields	Notes
HasReactionAliasOf	idx0		alias	This index can be used to find a particular alias when the type is not known.
HasReactionAliasOf	idxFrom		from-link, alias	This index allows the application to find a particular alias if the type is known.
HasReactionAliasOf	idxTo		to-link

HasRepresentativeOf Relationship

This relationship connects a genome to the FIGfam protein families for which it has representative proteins. This information can be computed from other relationships, but it is provided explicitly to allow fast access to a genome's FIGfam profile.

• Genome HasRepresentativeOf Family (many-to-many)
• Family IsRepresentedIn Genome (many-to-many)

HasRepresentativeOf Fields

Name	Type	Notes
from-link	string	id of the source Genome.
to-link	string	id of the target Family.

HasRepresentativeOf Indexes

Table	Name	Type	Fields	Notes
HasRepresentativeOf	idxFrom		from-link
HasRepresentativeOf	idxTo		to-link

HasResultsIn Relationship

This relationship connects a chip to the experiments that were applied to it.

• Chip HasResultsIn Experiment (one-to-many)
• Experiment HasResultsFor Chip (many-to-one)

HasResultsIn Fields

Name	Type	Notes
from-link	string	id of the source Chip.
sequence	int	Sequence number of this experiment in the various result vectors.
to-link	string	id of the target Experiment.

HasResultsIn Indexes

Table	Name	Type	Fields	Notes
HasResultsIn	idxFrom		from-link
HasResultsIn	idxTo	unique	to-link, sequence	This index allows you to access the experiments for a chip in sequence order.

HasSection Relationship

This relationship connects a contig to its actual DNA sequences.

• Contig HasSection DNASequence (one-to-many)
• DNASequence IsSectionOf Contig (many-to-one)

HasSection Fields

Name	Type	Notes
from-link	string	id of the source Contig.
to-link	string	id of the target DNASequence.

HasSection Indexes

Table	Name	Type	Fields	Notes
HasSection	idxFrom		from-link
HasSection	idxTo	unique	to-link

HasValueFor Relationship

This relationship connects an experiment to its attributes. The attribute values are stored here.

• Experiment HasValueFor Attribute (many-to-many)
• Attribute HasValueIn Experiment (many-to-many)

HasValueFor Fields

Name	Type	Notes
from-link	string	id of the source Experiment.
to-link	string	id of the target Attribute.
value	string	Value of this attribute in the given experiment. This is always encoded as a string, but may in fact be a number.

HasValueFor Indexes

Table	Name	Type	Fields	Notes
HasValueFor	idxFrom		from-link
HasValueFor	idxTo		to-link

HasValueOf Relationship

This relationship connects an alignment/tree project to a type of attribute that helps to describe the project. The value of the attribute is stored in the relationship itself.

• ATProject HasValueOf ATPAttribute (many-to-many)
• ATPAttribute IsSpecifiedFor ATProject (many-to-many)

HasValueOf Fields

Name	Type	Notes
from-link	string	id of the source ATProject.
to-link	string	id of the target ATPAttribute.
value	text	Value of this attribute for this project.

HasValueOf Indexes

Table	Name	Type	Fields	Notes
HasValueOf	idxFrom		from-link
HasValueOf	idxTo		to-link

Identifies Relationship

The normal case is that an identifier identifies a single feature, while a feature can have many identifiers. Some identifiers, however, are ambiguous and will connect to many features.

• Identifier Identifies Feature (many-to-many)
• Feature IsIdentifiedBy Identifier (many-to-many)

Identifies Fields

Name	Type	Notes
conf	char	Confidence code for this alias. "A" indicates it has been curated. other codes indicate less confidence.
from-link	string	id of the source Identifier.
to-link	string	id of the target Feature.

Identifies Indexes

Table	Name	Type	Fields	Notes
Identifies	idxFrom		from-link
Identifies	idxTo		to-link

Includes Relationship

A subsystem is defined by its roles. The subsystem's variants contain slightly different sets of roles, but all of the roles in a variant must be connected to the parent subsystem by this relationship. A subsystem always has at least one role, and a role always belongs to at least one subsystem.

• Subsystem Includes Role (many-to-many)
• Role IsIncludedIn Subsystem (many-to-many)

Includes Fields

Name	Type	Notes
abbreviation	string	Abbreviation for this role in this subsystem. The abbreviations are used in columnar displays, and they also appear on diagrams.
auxiliary	boolean	TRUE if this is an auxiliary role, or FALSE if this role is a functioning part of the subsystem.
from-link	string	id of the source Subsystem.
sequence	counter	Sequence number of the role within the subsystem. When the roles are formed into a variant, they will generally appear in sequence order.
to-link	string	id of the target Role.

Includes Indexes

Table	Name	Type	Fields	Notes
Includes	idxFrom		from-link, sequence	This index insures that the roles of the subsystem are presented in sequence order.
Includes	idxTo		to-link

IndicatedLevelsFor Relationship

This relationship connects a feature to a chip from which experimental data was produced for the feature. It contains a vector of the expression levels.

• Chip IndicatedLevelsFor Feature (many-to-many)
• Feature HasLevelsFrom Chip (many-to-many)

IndicatedLevelsFor Fields

Name	Type	Notes
from-link	string	id of the source Chip.
level-vector	countVector	Vector of expression levels (-1, 0, 1) for the experiments, in sequence order.
to-link	string	id of the target Feature.

IndicatedLevelsFor Indexes

Table	Name	Type	Fields	Notes
IndicatedLevelsFor	idxFrom		from-link
IndicatedLevelsFor	idxTo		to-link

IndicatesSignalFor Relationship

This relationship connects an experiment to a feature. The feature expression levels inferred from the experimental results are stored here.

• Experiment IndicatesSignalFor Feature (many-to-many)
• Feature HasIndicatedSignalFrom Experiment (many-to-many)

IndicatesSignalFor Fields

Name	Type	Notes
from-link	string	id of the source Experiment.
level	int	Indication of whether the feature is expressed (1), not expressed (-1), or unknown (0).
rma-value	float	Normalized expression value for this feature under the experiment's conditions.
to-link	string	id of the target Feature.

IndicatesSignalFor Indexes

Table	Name	Type	Fields	Notes
IndicatesSignalFor	idxFrom		from-link
IndicatesSignalFor	idxTo		to-link

Involves Relationship

This relationship connects a reaction to the compounds that participate in it. A reaction involves many compounds, and a compound can be involved in many reactions. The relationship attributes indicate whether a compound is a product or substrate of the reaction, as well as its stoichiometry.

• Reaction Involves Compound (many-to-many)
• Compound IsInvolvedIn Reaction (many-to-many)

Involves Fields

Name	Type	Notes
compartment	char	This character indicates the compartment where the reactant is located. This allows us to handle transport reactions. e stands for extracellular, c stands for cytosol, and p stands for periplasm.
discriminator	int	A unique ID for this record. The discriminator does not provide any useful data, but it prevents identical records from being collapsed by sort process used to load the database from constituent files.
from-link	string	id of the source Reaction.
main	boolean	TRUE if this compound is one of the main participants in the reaction, else FALSE. It is permissible for none of the compounds in the reaction to be considered main, in which case this value would be FALSE for all of the relevant compounds.
product	boolean	TRUE if the compound is a product of the reaction, FALSE if it is a substrate. When a reaction is written on paper in chemical notation, the substrates are left of the arrow and the products are to the right. Sorting on this field will cause the substrates to appear first, followed by the products. If the reaction is reversible, then the notion of substrates and products is not intuitive; however, a value here of FALSE still puts the compound left of the arrow and a value of TRUE still puts it to the right.
stoichiometry	float	Number of molecules of the compound that participate in a single instance of the reaction. For example, if a reaction produces two water molecules, the stoichiometry of water for the reaction would be two. When a reaction is written on paper in chemical notation, the stoichiometry is the number next to the chemical formula of the compound.
to-link	string	id of the target Compound.

Involves Indexes

Table	Name	Type	Fields	Notes
Involves	idxFrom		from-link
Involves	idxTo		to-link, product, main DESC	This index presents the compounds in the reaction in the order they should be displayed when writing it in chemical notation. All the substrates appear before all the products, and within that ordering, the main compounds appear first.

IsAnnotatedBy Relationship

This relationship connects a feature to its annotations. A feature may have multiple annotations, but an annotation belongs to only one feature.

• Feature IsAnnotatedBy Annotation (one-to-many)
• Annotation Annotates Feature (many-to-one)

IsAnnotatedBy Fields

Name	Type	Notes
from-link	string	id of the source Feature.
to-link	string	id of the target Annotation.

IsAnnotatedBy Indexes

Table	Name	Type	Fields	Notes
IsAnnotatedBy	idxFrom		from-link
IsAnnotatedBy	idxTo	unique	to-link

IsAttachmentSiteFor Relationship

This relationship connects an attachment site to the feature it attaches. This is an old table that will probably be replaced by the regulon-related entities and relationships.

• Feature IsAttachmentSiteFor Feature (many-to-many)
• Feature HasAttachmentSite Feature (many-to-many)

IsAttachmentSiteFor Fields

Name	Type	Notes
edge	char	Edge of the attached feature relevant to the site: L for the left edge, R for the right.
from-link	string	id of the source Feature.
to-link	string	id of the target Feature.

IsAttachmentSiteFor Indexes

Table	Name	Type	Fields	Notes
IsAttachmentSiteFor	idxFrom		from-link
IsAttachmentSiteFor	idxTo		to-link

IsCategorizedInto Relationship

This relationship connects an EC number to reactions that are consistent with the chemistry described by the EC class.

• Reaction IsCategorizedInto EcNumber (many-to-many)
• EcNumber Categorizes Reaction (many-to-many)

IsCategorizedInto Fields

Name	Type	Notes
from-link	string	id of the source Reaction.
source	string	This is the source of the EC number mapping for this reaction.
to-link	string	id of the target EcNumber.

IsCategorizedInto Indexes

Table	Name	Type	Fields	Notes
IsCategorizedInto	idxFrom		from-link
IsCategorizedInto	idxTo		to-link

IsClassFor Relationship

This relationship connects each subsystem class with the subsystems that belong to it. A class can contain many subsystems, but a subsystem is only in one class. Some subsystems are not in any class, but this is usually a temporary condition.

• SubsystemClass IsClassFor Subsystem (one-to-many)
• Subsystem IsInClass SubsystemClass (many-to-one)

IsClassFor Fields

Name	Type	Notes
from-link	string	id of the source SubsystemClass.
to-link	string	id of the target Subsystem.

IsClassFor Indexes

Table	Name	Type	Fields	Notes
IsClassFor	idxFrom		from-link
IsClassFor	idxTo	unique	to-link

IsCollectionOf Relationship

A genome belongs to only one genome set. For each set, this relationship marks the genome to be used as its representative.

• GenomeSet IsCollectionOf Genome (one-to-many)
• Genome IsCollectedInto GenomeSet (many-to-one)

IsCollectionOf Fields

Name	Type	Notes
from-link	string	id of the source GenomeSet.
representative	boolean	TRUE for the representative genome of the set, else FALSE.
to-link	string	id of the target Genome.

IsCollectionOf Indexes

Table	Name	Type	Fields	Notes
IsCollectionOf	idxFrom		from-link
IsCollectionOf	idxTo	unique	to-link

IsCombinationOf Relationship

This relationship combines roles into role sets. Each role set is a combination of roles that can trigger a reaction.

• RoleSet IsCombinationOf Role (many-to-many)
• Role IsInCombination RoleSet (many-to-many)

IsCombinationOf Fields

Name	Type	Notes
from-link	hash-string	id of the source RoleSet.
local	boolean	If a role combination is local, it is only applied if the genes are colocalized on the chromosome.
to-link	string	id of the target Role.

IsCombinationOf Indexes

Table	Name	Type	Fields	Notes
IsCombinationOf	idxFrom		from-link
IsCombinationOf	idxTo		to-link

IsConfiguredBy Relationship

This relationship connects a genome to the atomic regulons that describe its state.

• Genome IsConfiguredBy AtomicRegulon (one-to-many)
• AtomicRegulon ReflectsStateOf Genome (many-to-one)

IsConfiguredBy Fields

Name	Type	Notes
from-link	string	id of the source Genome.
to-link	string	id of the target AtomicRegulon.

IsConfiguredBy Indexes

Table	Name	Type	Fields	Notes
IsConfiguredBy	idxFrom		from-link
IsConfiguredBy	idxTo	unique	to-link

IsConsistentWith Relationship

This relationship connects a functional role to the EC numbers consistent with the chemistry described in the role.

• EcNumber IsConsistentWith Role (many-to-many)
• Role IsConsistentTo EcNumber (many-to-many)

IsConsistentWith Fields

Name	Type	Notes
from-link	string	id of the source EcNumber.
to-link	string	id of the target Role.

IsConsistentWith Indexes

Table	Name	Type	Fields	Notes
IsConsistentWith	idxFrom		from-link
IsConsistentWith	idxTo		to-link

IsCoregulatedWith Relationship

This relationship connects a feature with another feature in the same genome with which it appears to be coregulated as a result of expression data analysis.

• Feature IsCoregulatedWith Feature (many-to-many)
• Feature HasCoregulationWith Feature (many-to-many)

IsCoregulatedWith Fields

Name	Type	Notes
coefficient	float	Pearson correlation coefficient for this coregulation.
from-link	string	id of the source Feature.
to-link	string	id of the target Feature.

IsCoregulatedWith Indexes

Table	Name	Type	Fields	Notes
IsCoregulatedWith	idxFrom		from-link
IsCoregulatedWith	idxTo		to-link

IsCoupledTo Relationship

This relationship connects two FIGfams that we believe to be related either because their members occur in proximity on chromosomes or because the members are expressed together. Such a relationship is evidence the functions of the FIGfams are themselves related. This relationship is commutative; only the instance in which the first FIGfam has a lower ID than the second is stored.

• Family IsCoupledTo Family (many-to-many)
• Family IsCoupledWith Family (many-to-many)

IsCoupledTo Fields

Name	Type	Notes
co-expression-evidence	counter	number of times members of the two FIGfams are co-expressed in expression data experiments
co-occurrence-evidence	counter	number of times members of the two FIGfams occur close to each other on chromosomes
from-link	string	id of the source Family.
to-link	string	id of the target Family.

IsCoupledTo Indexes

Table	Name	Type	Fields	Notes
IsCoupledTo	idxFrom		from-link
IsCoupledTo	idxTo		to-link

IsDeterminedBy Relationship

A functional coupling evidence set exists because it has pairings in it, and this relationship connects the evidence set to its constituent pairings. A pairing cam belong to multiple evidence sets.

• PairSet IsDeterminedBy Pairing (one-to-many)
• Pairing Determines PairSet (many-to-one)

IsDeterminedBy Fields

Name	Type	Notes
from-link	int	id of the source PairSet.
inverted	boolean	A pairing is an unordered pair of protein sequences, but its similarity to other pairings in a pair set is ordered. Let (A,B) be a pairing and (X,Y) be another pairing in the same set. If this flag is FALSE, then (A =~ X) and (B =~ Y). If this flag is TRUE, then (A =~ Y) and (B =~ X).
to-link	string	id of the target Pairing.

IsDeterminedBy Indexes

Table	Name	Type	Fields	Notes
IsDeterminedBy	idxFrom		from-link
IsDeterminedBy	idxTo	unique	to-link

IsFamilyFor Relationship

This relationship connects an isofunctional family to the roles that make up its assigned function.

• Family IsFamilyFor Role (many-to-many)
• Role DeterminesFunctionOf Family (many-to-many)

IsFamilyFor Fields

Name	Type	Notes
from-link	string	id of the source Family.
to-link	string	id of the target Role.

IsFamilyFor Indexes

Table	Name	Type	Fields	Notes
IsFamilyFor	idxFrom		from-link
IsFamilyFor	idxTo		to-link

IsFormedOf Relationship

This relationship connects each feature to the atomic regulon to which it belongs.

• AtomicRegulon IsFormedOf Feature (one-to-many)
• Feature IsFormedInto AtomicRegulon (many-to-one)

IsFormedOf Fields

Name	Type	Notes
from-link	string	id of the source AtomicRegulon.
to-link	string	id of the target Feature.

IsFormedOf Indexes

Table	Name	Type	Fields	Notes
IsFormedOf	idxFrom		from-link
IsFormedOf	idxTo	unique	to-link

IsFoundIn Relationship

This relationship connects a compound to the known structural cues. The structural cues enable the application to query compounds by molecular structure.

• StructuralCue IsFoundIn Compound (many-to-many)
• Compound IsComposedOf StructuralCue (many-to-many)

IsFoundIn Fields

Name	Type	Notes
count	int	Number of times the cue occurs in the compound.
from-link	string	id of the source StructuralCue.
to-link	string	id of the target Compound.

IsFoundIn Indexes

Table	Name	Type	Fields	Notes
IsFoundIn	idxFrom		from-link
IsFoundIn	idxTo		to-link

IsFunctionalIn Relationship

This relationship connects a role with the features in which it plays a functional part. In most cases, the functional assignment of a feature is also its role; however, some features implement multiple roles, and each individual role is connected by a separate instance of this relationship.

• Role IsFunctionalIn Feature (many-to-many)
• Feature HasFunctional Role (many-to-many)

IsFunctionalIn Fields

Name	Type	Notes
from-link	string	id of the source Role.
to-link	string	id of the target Feature.

IsFunctionalIn Indexes

Table	Name	Type	Fields	Notes
IsFunctionalIn	idxFrom		from-link
IsFunctionalIn	idxTo		to-link

IsGroupFor Relationship

The recursive IsGroupContaining relationship organizes taxonomic groupings into a hierarchy based on the standard organism taxonomy.

• TaxonomicGrouping IsGroupFor TaxonomicGrouping (one-to-many)
• TaxonomicGrouping IsInGroup TaxonomicGrouping (many-to-one)

IsGroupFor Fields

Name	Type	Notes
from-link	counter	id of the source TaxonomicGrouping.
to-link	counter	id of the target TaxonomicGrouping.

IsGroupFor Indexes

Table	Name	Type	Fields	Notes
IsGroupFor	idxFrom		from-link
IsGroupFor	idxTo	unique	to-link

IsImplementedBy Relationship

This relationship connects a variant to the physical machines that implement it in the genomes. A variant is implemented by many machines, but a machine belongs to only one variant.

• Variant IsImplementedBy MolecularMachine (one-to-many)
• MolecularMachine Implements Variant (many-to-one)

IsImplementedBy Fields

Name	Type	Notes
from-link	hash-string	id of the source Variant.
to-link	hash-string	id of the target MolecularMachine.

IsImplementedBy Indexes

Table	Name	Type	Fields	Notes
IsImplementedBy	idxFrom		from-link
IsImplementedBy	idxTo	unique	to-link

IsInPair Relationship

A pairing contains exactly two protein sequences. A protein sequence can belong to multiple pairings. When going from a protein sequence to its pairings, they are presented in alphabetical order by sequence key.

• Feature IsInPair Pairing (many-to-many)
• Pairing IsPairOf Feature (many-to-many)

IsInPair Fields

Name	Type	Notes
from-link	string	id of the source Feature.
to-link	string	id of the target Pairing.

IsInPair Indexes

Table	Name	Type	Fields	Notes
IsInPair	idxFrom		from-link
IsInPair	idxTo		to-link

IsLocatedIn Relationship

A feature is a set of DNA sequence fragments. Most features are a single contiquous fragment, so they are located in only one DNA sequence; however, fragments have a maximum length, so even a single contiguous feature may participate in this relationship multiple times. A few features belong to multiple DNA sequences. In that case, however, all the DNA sequences belong to the same genome. A DNA sequence itself will frequently have thousands of features connected to it.

• Feature IsLocatedIn Contig (many-to-many)
• Contig IsLocusFor Feature (many-to-many)

IsLocatedIn Fields

Name	Type	Notes
begin	int	Index (1-based) of the first residue in the contig that belongs to the segment.
dir	char	Direction (strand) of the segment: "+" if it is forward and "-" if it is backward.
from-link	string	id of the source Feature.
len	int	Length of this segment.
ordinal	int	Sequence number of this segment, starting from 1 and proceeding sequentially forward from there.
to-link	string	id of the target Contig.

IsLocatedIn Indexes

Table	Name	Type	Fields	Notes
IsLocatedIn	idxFrom		from-link, ordinal	This index allows the application to find all the segments of a feature in the proper order.
IsLocatedIn	idxTo		to-link, begin	This index is the one used by applications to find all the feature segments that contain a specific residue.

IsMachineOf Relationship

This relationship connects a molecular machine to its various machine roles. Each machine has many machine roles, but each machine role belongs to only one machine.

• MolecularMachine IsMachineOf MachineRole (one-to-many)
• MachineRole IsRoleFor MolecularMachine (many-to-one)

IsMachineOf Fields

Name	Type	Notes
from-link	hash-string	id of the source MolecularMachine.
to-link	string	id of the target MachineRole.

IsMachineOf Indexes

Table	Name	Type	Fields	Notes
IsMachineOf	idxFrom		from-link
IsMachineOf	idxTo	unique	to-link

IsMadeUpOf Relationship

This relationship connects each genome to the DNA sequences that make it up.

• Genome IsMadeUpOf Contig (one-to-many)
• Contig MakesUp Genome (many-to-one)

IsMadeUpOf Fields

Name	Type	Notes
from-link	string	id of the source Genome.
to-link	string	id of the target Contig.

IsMadeUpOf Indexes

Table	Name	Type	Fields	Notes
IsMadeUpOf	idxFrom		from-link
IsMadeUpOf	idxTo	unique	to-link

IsModeledBy Relationship

A genome can be modeled by many different models, but a model belongs to only one genome.

• Genome IsModeledBy Model (one-to-many)
• Model Models Genome (many-to-one)

IsModeledBy Fields

Name	Type	Notes
from-link	string	id of the source Genome.
to-link	string	id of the target Model.

IsModeledBy Indexes

Table	Name	Type	Fields	Notes
IsModeledBy	idxFrom		from-link
IsModeledBy	idxTo	unique	to-link

IsNamedBy Relationship

The normal case is that an identifier names a single protein sequence, while a protein sequence can have many identifiers, but some identifiers name multiple sequences. This relationship is populated with data submitted to the annotation clearinghouse as well as external IDs in the non-redundant database.

• ProteinSequence IsNamedBy Identifier (many-to-many)
• Identifier Names ProteinSequence (many-to-many)

IsNamedBy Fields

Name	Type	Notes
from-link	string	id of the source ProteinSequence.
to-link	string	id of the target Identifier.

IsNamedBy Indexes

Table	Name	Type	Fields	Notes
IsNamedBy	idxFrom		from-link
IsNamedBy	idxTo		to-link

IsOrganizedBy Relationship

This relationship connects each tree to the alignment it organizes. Each tree belongs to only one alignment.

• Alignment IsOrganizedBy AlignmentTree (one-to-many)
• AlignmentTree Organizes Alignment (many-to-one)

IsOrganizedBy Fields

Name	Type	Notes
from-link	int	id of the source Alignment.
to-link	int	id of the target AlignmentTree.

IsOrganizedBy Indexes

Table	Name	Type	Fields	Notes
IsOrganizedBy	idxFrom		from-link
IsOrganizedBy	idxTo	unique	to-link

IsOwnerOf Relationship

This relationship connects each feature to its parent genome.

• Genome IsOwnerOf Feature (one-to-many)
• Feature IsOwnedBy Genome (many-to-one)

IsOwnerOf Fields

Name	Type	Notes
from-link	string	id of the source Genome.
to-link	string	id of the target Feature.

IsOwnerOf Indexes

Table	Name	Type	Fields	Notes
IsOwnerOf	idxFrom		from-link
IsOwnerOf	idxTo	unique	to-link

IsPresentIn Relationship

This relationship connects a media to the compounds present in it.

• Compound IsPresentIn Media (many-to-many)
• Media IsComprisedOf Compound (many-to-many)

IsPresentIn Fields

Name	Type	Notes
concentration	float	Concentration of the compound in the media, expressed in mM/liter.
from-link	string	id of the source Compound.
maximum-uptake	float	Upper limit on consumption of compound by cells in the media, expressed in mM/gm CDW hr.
to-link	string	id of the target Media.

IsPresentIn Indexes

Table	Name	Type	Fields	Notes
IsPresentIn	idxFrom		from-link
IsPresentIn	idxTo		to-link

IsProjectFor Relationship

This relationship connects an alignment to the project for which it was generated. Each alignment belongs to only one project.

• ATProject IsProjectFor Alignment (one-to-many)
• Alignment IsPartOf ATProject (many-to-one)

IsProjectFor Fields

Name	Type	Notes
from-link	string	id of the source ATProject.
to-link	int	id of the target Alignment.

IsProjectFor Indexes

Table	Name	Type	Fields	Notes
IsProjectFor	idxFrom		from-link
IsProjectFor	idxTo	unique	to-link

IsProteinFor Relationship

This relationship connects a peg feature to the protein sequence it produces (if any). Only peg features participate in this relationship. A single protein sequence will frequently be produced by many features.

• ProteinSequence IsProteinFor Feature (one-to-many)
• Feature Produces ProteinSequence (many-to-one)

IsProteinFor Fields

Name	Type	Notes
from-link	string	id of the source ProteinSequence.
to-link	string	id of the target Feature.

IsProteinFor Indexes

Table	Name	Type	Fields	Notes
IsProteinFor	idxFrom		from-link
IsProteinFor	idxTo	unique	to-link

IsRegulatedWith Relationship

This relationship connects a feature to the set of coregulated features.

• Feature IsRegulatedWith CoregulatedSet (many-to-many)
• CoregulatedSet Regulates Feature (many-to-many)

IsRegulatedWith Fields

Name	Type	Notes
from-link	string	id of the source Feature.
to-link	hash-string	id of the target CoregulatedSet.

IsRegulatedWith Indexes

Table	Name	Type	Fields	Notes
IsRegulatedWith	idxFrom		from-link
IsRegulatedWith	idxTo		to-link

IsRelevantFor Relationship

This relationship connects a diagram to the subsystems that are depicted on it. Only diagrams which are useful in curating or annotation the subsystem are specified in this relationship.

• Diagram IsRelevantFor Subsystem (many-to-many)
• Subsystem IsRelevantTo Diagram (many-to-many)

IsRelevantFor Fields

Name	Type	Notes
from-link	string	id of the source Diagram.
to-link	string	id of the target Subsystem.

IsRelevantFor Indexes

Table	Name	Type	Fields	Notes
IsRelevantFor	idxFrom		from-link
IsRelevantFor	idxTo		to-link

IsRequiredBy Relationship

This relationship specifies the reactions that occur in the model. The direction is an attribute of this relationship.

• Reaction IsRequiredBy Model (many-to-many)
• Model Requires Reaction (many-to-many)

IsRequiredBy Fields

Name	Type	Notes
class	char	This describes the behavior of the reaction in complete media. The possible values are 'b' blocked, 'p' positive, 'n' negative, 'v' variable, 'd' dead, 'e' negative variable, and 'o' positive variable.
compartment	char	This field indicates the compartment the reaction is taking place in. Possible values are 'e' extracellular, 'c' cytosol, and 'p' perisplasm.
confidence	int	This is the level of confidence that this reaction is active in this model. '1' is the highest confidence and '10' is the lowest confidence.
dependency	text	This strings holds data about why the reaction is included in the model.
direction	char	This field indicates the direction in which the reaction operates in the context of the model. The values are 'F' (forward), 'B' (backward) and 'R' (reversible).
from-link	string	id of the source Reaction.
notes	text	This string holds information about the reaction in the model.
to-link	string	id of the target Model.

IsRequiredBy Indexes

Table	Name	Type	Fields	Notes
IsRequiredBy	idxFrom		from-link
IsRequiredBy	idxTo		to-link

IsRoleOf Relationship

This relationship connects a role to the machine roles that represent its appearance in a molecular machine. A machine role has exactly one associated role, but a role may be represented by many machine roles.

• Role IsRoleOf MachineRole (one-to-many)
• MachineRole HasRole Role (many-to-one)

IsRoleOf Fields

Name	Type	Notes
from-link	string	id of the source Role.
to-link	string	id of the target MachineRole.

IsRoleOf Indexes

Table	Name	Type	Fields	Notes
IsRoleOf	idxFrom		from-link
IsRoleOf	idxTo	unique	to-link

IsSelectedBy Relationship

This relationship connects an alignment/tree project to the roles used to select the protein sequences aligned.

• Role IsSelectedBy ATProject (many-to-many)
• ATProject IsSelectedWith Role (many-to-many)

IsSelectedBy Fields

Name	Type	Notes
from-link	string	id of the source Role.
to-link	string	id of the target ATProject.

IsSelectedBy Indexes

Table	Name	Type	Fields	Notes
IsSelectedBy	idxFrom		from-link
IsSelectedBy	idxTo		to-link

IsSetOf Relationship

This relationship connects a feature set to its constituent features. All of the features in a set must be present for a set to stimulate a reaction.

• FeatureSet IsSetOf Feature (many-to-many)
• Feature IsInSet FeatureSet (many-to-many)

IsSetOf Fields

Name	Type	Notes
from-link	hash-string	id of the source FeatureSet.
to-link	string	id of the target Feature.

IsSetOf Indexes

Table	Name	Type	Fields	Notes
IsSetOf	idxFrom		from-link
IsSetOf	idxTo		to-link

IsStimulatedBy Relationship

In the context of a model, a reaction is stimulated when all features in a feature set are present. By comparing feature sets to role sets we can determine whether a model is consistent with annotations.

• Reaction IsStimulatedBy FeatureSet (many-to-many)
• FeatureSet Stimulates Reaction (many-to-many)

IsStimulatedBy Fields

Name	Type	Notes
from-link	string	id of the source Reaction.
to-link	hash-string	id of the target FeatureSet.

IsStimulatedBy Indexes

Table	Name	Type	Fields	Notes
IsStimulatedBy	idxFrom		from-link
IsStimulatedBy	idxTo		to-link

IsSubInstanceOf Relationship

This relationship connects a scenario to its subsystem it validates. A scenario belongs to exactly one subsystem, but a subsystem may have multiple scenarios.

• Subsystem IsSubInstanceOf Scenario (one-to-many)
• Scenario Validates Subsystem (many-to-one)

IsSubInstanceOf Fields

Name	Type	Notes
from-link	string	id of the source Subsystem.
to-link	int	id of the target Scenario.

IsSubInstanceOf Indexes

Table	Name	Type	Fields	Notes
IsSubInstanceOf	idxFrom		from-link
IsSubInstanceOf	idxTo	unique	to-link

IsSuperclassOf Relationship

This is a recursive relationship that imposes a hierarchy on the subsystem classes.

• SubsystemClass IsSuperclassOf SubsystemClass (one-to-many)
• SubsystemClass IsSubclassOf SubsystemClass (many-to-one)

IsSuperclassOf Fields

Name	Type	Notes
from-link	string	id of the source SubsystemClass.
to-link	string	id of the target SubsystemClass.

IsSuperclassOf Indexes

Table	Name	Type	Fields	Notes
IsSuperclassOf	idxFrom		from-link
IsSuperclassOf	idxTo	unique	to-link

IsTaxonomyOf Relationship

A genome is assigned to a particular point in the taxonomy tree, but not necessarily to a leaf node. In some cases, the exact species and strain is not available when inserting the genome, so it is placed at the lowest node that probably contains the actual genome.

• TaxonomicGrouping IsTaxonomyOf Genome (one-to-many)
• Genome IsInTaxa TaxonomicGrouping (many-to-one)

IsTaxonomyOf Fields

Name	Type	Notes
from-link	counter	id of the source TaxonomicGrouping.
to-link	string	id of the target Genome.

IsTaxonomyOf Indexes

Table	Name	Type	Fields	Notes
IsTaxonomyOf	idxFrom		from-link
IsTaxonomyOf	idxTo	unique	to-link

IsTerminusFor Relationship

A terminus for a scenario is a compound that acts as its input or output. A compound can be the terminus for many scenarios, and a scenario will have many termini. The relationship attributes indicate whether the compound is an input to the scenario or an output. In some cases, there may be multiple alternative output groups. This is also indicated by the attributes.

• Compound IsTerminusFor Scenario (many-to-many)
• Scenario HasAsTerminus Compound (many-to-many)

IsTerminusFor Fields

Name	Type	Notes
from-link	string	id of the source Compound.
group-number	int	If zero, then the compound is an input. If one, the compound is an output. If two, the compound is an auxiliary output.
to-link	int	id of the target Scenario.

IsTerminusFor Indexes

Table	Name	Type	Fields	Notes
IsTerminusFor	idxFrom		from-link
IsTerminusFor	idxTo		to-link, group-number	This index allows the application to view a scenario's compounds by group.

IsTriggeredBy Relationship

A reaction can be triggered by many role sets. A role set can trigger many reactions.

• Reaction IsTriggeredBy RoleSet (many-to-many)
• RoleSet Triggers Reaction (many-to-many)

IsTriggeredBy Fields

Name	Type	Notes
from-link	string	id of the source Reaction.
master	boolean	This boolean allows us to turn on and off certain mappings. If this boolean is true, the roleset will trigger the addition of this reaction in the model building process.
source	string	This is the source of the mapping between the reaction and role set.
subsystem	text	This is the subsystem associated with this reaction and roleset.
to-link	hash-string	id of the target RoleSet.

IsTriggeredBy Indexes

Table	Name	Type	Fields	Notes
IsTriggeredBy	idxFrom		from-link
IsTriggeredBy	idxTo		to-link

OccursIn Relationship

This relationship connects features to clusters. It is generally expected that the presence of features in a cluster indicates a functional relationship.

• Feature OccursIn Cluster (many-to-many)
• Cluster IsOccurrenceOf Feature (many-to-many)

OccursIn Fields

Name	Type	Notes
from-link	string	id of the source Feature.
to-link	int	id of the target Cluster.

OccursIn Indexes

Table	Name	Type	Fields	Notes
OccursIn	idxFrom		from-link
OccursIn	idxTo		to-link

OperatesIn Relationship

This relationship connects an experiment to the media in which the experiment took place.

• Experiment OperatesIn Media (many-to-many)
• Media IsUtilizedIn Experiment (many-to-many)

OperatesIn Fields

Name	Type	Notes
from-link	string	id of the source Experiment.
to-link	string	id of the target Media.

OperatesIn Indexes

Table	Name	Type	Fields	Notes
OperatesIn	idxFrom		from-link
OperatesIn	idxTo		to-link

Overlaps Relationship

A Scenario overlaps a diagram when the diagram displays a portion of the reactions that make up the scenario. A scenario may overlap many diagrams, and a diagram may be include portions of many scenarios.

• Scenario Overlaps Diagram (many-to-many)
• Diagram IncludesPartOf Scenario (many-to-many)

Overlaps Fields

Name	Type	Notes
from-link	int	id of the source Scenario.
to-link	string	id of the target Diagram.

Overlaps Indexes

Table	Name	Type	Fields	Notes
Overlaps	idxFrom		from-link
Overlaps	idxTo		to-link

ProducedResultsFor Relationship

This relationship connects a chip to a genome for which it was used to produce experimental results. In general, a chip is used for only one genome and vice versa, but this is not a requirement.

• Chip ProducedResultsFor Genome (many-to-many)
• Genome HadResultsProducedBy Chip (many-to-many)

ProducedResultsFor Fields

Name	Type	Notes
from-link	string	id of the source Chip.
to-link	string	id of the target Genome.

ProducedResultsFor Indexes

Table	Name	Type	Fields	Notes
ProducedResultsFor	idxFrom		from-link
ProducedResultsFor	idxTo		to-link

Shows Relationship

This relationship indicates that a compound appears on a particular diagram. The same compound can appear on many diagrams, and a diagram always contains many compounds.

• Diagram Shows Compound (many-to-many)
• Compound IsShownOn Diagram (many-to-many)

Shows Fields

Name	Type	Notes
from-link	string	id of the source Diagram.
location	rectangle	Location of the compound's node on the diagram.
to-link	string	id of the target Compound.

Shows Indexes

Table	Name	Type	Fields	Notes
Shows	idxFrom		from-link
Shows	idxTo		to-link

Uses Relationship

This relationship connects a genome to the machines that form its metabolic pathways. A genome can use many machines, but a machine is used by exactly one genome.

• Genome Uses MolecularMachine (one-to-many)
• MolecularMachine IsUsedBy Genome (many-to-one)

Uses Fields

Name	Type	Notes
from-link	string	id of the source Genome.
to-link	hash-string	id of the target MolecularMachine.

Uses Indexes

Table	Name	Type	Fields	Notes
Uses	idxFrom		from-link
Uses	idxTo	unique	to-link

UsesProteinsIn Relationship

This relationship connects a project to the representative sets that comprise the sequences in its alignments. Each alignment in the project must correspond to one of the sets associated with the project.

• ATProject UsesProteinsIn RepresentativeSet (many-to-many)
• RepresentativeSet ContainsProteinsUsedBy ATProject (many-to-many)

UsesProteinsIn Fields

Name	Type	Notes
from-link	string	id of the source ATProject.
to-link	hash-string	id of the target RepresentativeSet.

UsesProteinsIn Indexes

Table	Name	Type	Fields	Notes
UsesProteinsIn	idxFrom		from-link
UsesProteinsIn	idxTo		to-link

BelongsTo Shape

This relationship is not physically implemented in the database. It is implicit in the data for a variant. A variant contains a boolean expression that describes the various combinations of roles it can contain.

The Sapling database is a distributable, self-contained copy of the NMPDR data. Unlike Sprout, which is optimized for searching, Sapling is designed to be structurally simple without sacrificing the ability to find information quickly.

The diagram colors indicate the way the data is loaded: each color indicates a different load group. In this version of Sapling everything but the Drug group (purple) is available in the database.

• Red Genome group: includes all taxonomy and sequence data.
• Blue Subsystem group: includes the subsystems themselves and all the subsystem spreadsheets.
• Green Feature group: includes features, feature identifiers, protein sequences, and related publications.
• Yellow Scenario group: includes KEGG diagrams, scenarios, and connections to the related chemical reactions.
• Purple Alignment group: includes feature alignment and tree projects.
• Navy Model group: includes compounds, reactions, structural cues, and models for the cell chemistry.
• Brown Family group: includes FIGfams and functional coupling data.
• Black Protein group: includes annotation history and data from the Annotation Clearinghouse.
• Cyan Expression group: includes data relating to gene expression and cell regulation.