Quaesia database description

MeSH Concept Schema (cMESH.json)

This table describes the structure of the entities, likely derived from MeSH (Medical Subject Headings), as they are stored in the database.
The cMESH integrates a total of ~355,000 MeSH records, structured as follows:

Descriptors: Approximately 30,000 official MeSH descriptors
Supplementary Concepts: Approximately 325,000 supplementary concept records

Each supplementary concept record is explicitly linked to its closest fitting Descriptor, which serves as its direct parent in the database hierarchy.
Note on Tree Number: To maintain hierarchical structure for Supplementary Concepts (which do not have native Tree Numbers), a Tree Number is synthesized. This is achieved by taking the parent's Tree Number and concatenating it with a dot (.) and the MeSH Identifier (e.g. D05.750.716.822.111.C472990).

Code	Description	Data type
_id	Custom MeSH ID (cMESHId): The primary key. This is a custom integer representation derived from the original MeSH Unique Identifier (UI) for efficient indexing. Encoding: UI leading 'C' → 3; UI leading 'D' → 4 (e.g. "C030544" → 3030544)	Integer
p	Preferred Name: The official or primary name for the MeSH concept/descriptor	String
d	Description: The text definition or scope note for the MeSH concept	String
lTN	Tree Numbers: A mapping of MeSH Tree Numbers to the count of direct children beneath that specific branch	JSON Object ({String:Integer...})
lC	Direct Children: A list of the cMESHId for all direct child concepts/nodes in the MeSH hierarchy	Array ([Integer...])
lt	Synonyms/Alternate Terms: A list of alternative terms associated with the concept	Array ([String...])

Interaction Concept Schema (cIT.json)

This table describes the controlled vocabulary used by Quaesia for classifying interaction types between entities.

Code	Description	Data type
_id	Custom Interaction Term ID (cITId): The primary key. An ad hoc integer identifier used internally by Quaesia to represent a specific interaction concept	Integer
p	Preferred Name: the common name used to describe this interaction concept (e.g., "negative", "upstream", "activated/caused")	String
l	Associated Terms: A list of synonyms or related phrases used to identify this interaction concept	Array ([String...])
c	Category: The broad classification or group to which this interaction concept belongs	String

Publication Schema (cPMID.json)

This table stores metadata and structural data for biomedical scientific publications, indexed by their PubMed ID (PMID).

Code	Description	Data type
_id	PubMed ID (PMID): The unique identifier for the publication, serving as the primary key	Integer
jt	Journal Title: The official title of the journal where the article was published	String
y	Publication Year: The year the article was published	Integer
lS	Section Sentences: A list of text sections and their corresponding sentence identifiers (cSentId). The section name indicates the type: "T" → Title "A" → Abstract otherwise author-defined section name	Array of Tuples [(String, [Float...])...])
lA	Authors: A list of authors identified by their Custom Author ID (cAuthorsId) paired with the index number corresponding to their affiliation in the laff list
laff	Affiliations: A list of all unique affiliation strings for the authors in this publication	String...]
PMC	PMC ID: The PubMed Central unique identifier	Integer
doi	DOI Link: The Digital Object Identifier link (may include the prefix)	String

Sentence Schema (cSent.json)

This table stores individual, normalized sentences extracted from biomedical publications.
Note on Normalization: Sentences have been normalized (e.g., removing double spaces and HTML tags). Consequently, the character-level positions (p) may not perfectly align with results from independent text analyses on the original, uncleaned source text.

Code	Description	Data type
_id	Custom Sentence ID (cSentId): The primary key. This is a composite ID calculated as PMID + (Sentence Index / 1,000,000)	Float
t	Text: The full, normalized text of the sentence	String
p	Start Position: The character-level starting index of the sentence within the concatenated publication text (Title + Abstract)	Integer

Author Schema (cAuthor.json)

Code	Description	Data type
_id	Custom Author ID (cAuthorsId): The primary key. A string identifier created by concatenating the author's First Name and Last Name (e.g., "FirstName+"\|"+LastName")	String
lp	Published PMIDs: A list of PubMed IDs (PMIDs) corresponding to publications authored by this person	Array ([Integer...])
lM	MeSH of Interest: A list of tuples containing a Custom MeSH ID (cMESHId) and the count of PMIDs published by this author that are associated with that specific MeSH term	Array of Tuples [(Integer, Integer)...]

MeSH List Schema (lcMESHId.json)

This table provides predefined, categorized lists of Custom MeSH IDs (cMESHId) for specific biomedical concepts.

Code	Description	Data type
_id	List Identifier: The unique name of the predefined list, indicating the category of MeSH concepts contained within. 0: All available cMESHIds MESH_CID: MeSH concepts mapped to a PubChem Compound ID (CID) MESH_FDAApproved: Chemical entities that are FDA-approved MESH_antibody: Chemical antibodies MESH_bodyPart_organ: body parts and organs MESH_cell_structure: cells and biological structures MESH_defect_deficiency_disorder: defects, deficiencies, and disorders MESH_disease: diseases MESH_human_protein: human proteins MESH_infection: infections MESH_phenomen_process: biological phenomena and processes MESH_side_effect: side effects MESH_sign_symptom: clinical signs and symptoms MESH_syndrome: syndromes MESH_tissue: biological tissue	Integer or String
l	Custom MeSH IDs (cMESHId): A list of the integer identifiers corresponding to the MeSH concepts for the specified category	Array ([Integer...])

Interaction Term List Schema (lcIT.json)

Code	Description	Data type
_id	List Identifier: The unique identifier or name of the predefined list. Predefined List Categories: 0: A list containing all available cITIds	Integer
l	Custom Interaction Term IDs: The list of integer identifiers (cITId) belonging to the category specified in the _id field	Array ([Integer...])

PubMed ID List Schema (lPMID.json)

This table organizes all PubMed IDs (PMIDs) indexed in the database based on the source PubMed XML file from the PubMed FTP server.
Note on Chunking: The total list was partitioned into sublists to comply with MongoDB's 16 MB BSON document size limit.

Code	Description	Data type
_id	Source File Index: The index number corresponding to the source PubMedXXXX.xml file retrieved from the PubMed FTP server	Integer
l	PubMed IDs (PMIDs): A list of all PMIDs sourced from the corresponding PubMedXXXX.xml file. Data Deduping Rule: If a PMID appears in multiple XML files, it is only included in the list associated with the highest source file index number	Array ([Integer...])

Sentence ID List Schema (lcSentId.json)

This table organizes all Custom Sentence IDs (cSentId) indexed in the database based on the source PubMed XML file from the PubMed FTP server.
Note on Chunking: The total list was partitioned into sublists to comply with MongoDB's 16 MB BSON document size limit.

Code	Description	Data type
_id	Source File Index: The index number corresponding to the source PubMedXXXX.xml file retrieved from the PubMed FTP server	Integer
l	Custom Sentence IDs (cSentId): A list of all sentence identifiers sourced from the corresponding PubMedXXXX.xml file. Data Deduping Rule: If a cSentId appears in multiple XML files, it is only included in the list associated with the highest source file index number	[Float...]

Author ID List Schema (lcAuthorId.json)

This table is used to chunk and organize all Custom Author IDs (cAuthorsId) into manageable sublists.
Note on Chunking: The total list was partitioned into sublists to comply with MongoDB's 16 MB BSON document size limit.

Code	Description	Data type
_id	List Index: An incremental integer number used to index and identify each chunk/sublist of author IDs	Integer
l	Author ID Sublist: A sublist containing Custom Author IDs (cAuthorsId). Each sublist is capped at a maximum of 10,000 author identifiers	Array ([String...])

Publication Reference Index Schema (cPMID_cite_cPMID.json)

This table provides an explicit, indexed list of PubMed ID (PMID) pairs where the first publication cites the second (Reference Index).

Code	Description	Data type
_id	Record Index: An incremental integer number used as a sequential index for the records in this mapping table	Integer
1	Citing PMID: The PubMed ID of the publication that includes the reference.	Integer
2	Cited PMID: The PubMed ID of the referenced publication.	Integer

Automatic Annotation Schema (cSent_2_autoAnno.json)

This table stores the automatic Named Entity Recognition and Disambiguation (NERD) and Relationship Extraction (RE) results for individual sentences.

Code	Description	Data type
_id	Custom Sentence ID (cSentId): The primary key. A composite identifier calculated as PMID + (Sentence Index / 1,000,000)	Float
N	Entity Detection: A list of all entities automatically detected within the sentence	Array of Objects
	Start: Character-level start position of the entity within the sentence	Integer
	End: Character-level end position of the entity within the sentence	Integer
	Associated cMESHIds: List of Custom MeSH IDs (cMESHId) associated with the detected entity	Array ([Integer...])
R	Relationship Extraction: A list of relationships automatically extracted between pairs of entities in the sentence	Array of Objects
	Start of entity 1: Character-level start position of the first entity in the pair	Integer
	End of entity 1: Character-level end position of the first entity in the pair	Integer
	Start of entity 2: Character-level start position of the second entity in the pair	Integer
	End of entity 2: Character-level end position of the second entity in the pair	Integer
	List of associated cITId (cITId, score): A list of tuples, where each tuple contains a Custom Interaction Term ID (cITId) and the confidence score for that specific relationship type	Array of Tuples ([(Integer, Integer)...])

MeSH-to-Publication Index Schema (cMESHId_citeNERD_2lPMID.json)

This table serves as a reverse index, listing all PubMed IDs (PMIDs) where a specific Custom MeSH ID (cMESHId) has been detected by the Named Entity Recognition and Disambiguation (NERD) system.

Code	Description	Data type
_id	Custom MeSH ID (cMESHId): The primary key. The integer identifier of the MeSH concept whose occurrences are being indexed	Integer
l	Publications List (PMIDs): A list of PubMed IDs where the entity corresponding to the cMESHId was identified in the text	Array ([Integer...])

Note on MongoDB Limit: Some PMID lists exceed the 16 MB limit. To successfully load this data into a MongoDB system, you will need to partition it into smaller documents (or 'chunks').

Relationship Evidence Sentence List Schema (cMESHId1_cMESHId2_cITId_lcSentId.json)

Code	Description	Data type
1	Custom MeSH ID 1 (cMESHId1): The first MeSH concept identifier in the relationship pair. Ordering Constraint: This ID is always the smaller value (cMESHId1 < cMESHId2) to ensure a unique key for the pair	Integer
2	Custom MeSH ID 2 (cMESHId2): The second MeSH concept identifier in the relationship pair. Ordering Constraint: This ID is always the larger value (cMESHId1 < cMESHId2)	Integer
t	Interaction Type ID (cITId): The Custom Interaction Term ID that defines the type of relationship found between cMESHId1 and cMESHId2.	Integer
l	Sentence ID List (cSentId): The list of Custom Sentence IDs where the relationship between this specific entity pair and interaction type was detected	[Float...]

Note on MongoDB Limit: Some cSentId lists exceed the 16 MB limit. To successfully load this data into a MongoDB system, you will need to partition it into smaller documents (or 'chunks').

Custom MeSH to ChEBI Mapping Schema (cMESHId_map_CHEBI.json)

This table provides a mapping between Quaesia's internal Custom MeSH Identifiers (cMESHId) and the external ChEBI (Chemical Entities of Biological Interest) Identifiers.

Code	Description	Data type
_id	Record Index: An incremental integer number used as a sequential index for the records in this mapping table	Integer
1	Custom MeSH ID (cMESHId): The integer identifier for the MeSH concept (often a chemical entity) being mapped	Integer
2	ChEBI Identifier (CHEBI ID): The unique integer ID assigned by the ChEBI database for the corresponding chemical entity	Integer

Custom MeSH to ChEMBL Mapping Schema (cMESHId_map_CHEMBL.json)

This table provides a mapping between Quaesia's internal Custom MeSH Identifiers (cMESHId) and the external ChEMBL (Chemical Database of Bioactive Molecules) Identifiers.

Code	Description	Data type
_id	Record Index: An incremental integer number used as a sequential index for the records in this mapping table	Integer
1	Custom MeSH ID (cMESHId): The integer identifier for the MeSH concept (often a chemical entity) being mapped	Integer
2	CHEMBL identifier (CHEMBL ID): The unique string ID assigned by the ChEMBL database for the corresponding bioactive molecule	String

Custom MeSH to Cell Line Mapping Schema (cMESHId_map_CL.json)

This table provides a mapping between Quaesia's internal Custom MeSH Identifiers (cMESHId) and the external Cell Line Ontology (CL) Identifiers.

Code	Description	Data type
_id	Record Index: An incremental integer number used as a sequential index for the records in this mapping table	Integer
1	Custom MeSH ID (cMESHId): The integer identifier for the MeSH concept related to a cell line or cell type being mapped	Integer
2	Cell Line Identifier (CL ID):The unique string identifier assigned by the external Cell Line Ontology	String

Custom MeSH to Ensembl Gene Mapping Schema (cMESHId_map_ENSEMBLG.json)

This table provides a mapping between Quaesia's internal Custom MeSH Identifiers (cMESHId) and the external NCBI Gene Identifiers (Entrez Gene IDs).

Code	Description	Data type
_id	Record Index: An incremental integer number used as a sequential index for the records in this mapping table	Integer
1	Custom MeSH ID (cMESHId): The integer identifier for the MeSH concept (often a gene or protein) being mapped	Integer
2	Ensembl Gene Identifier (ENSEMBLG): The unique string ID assigned by the Ensembl database for the corresponding gene	String

Custom MeSH to GeneId Mapping Schema (cMESHId_map_GENEID.json)

This table provides a mapping between Quaesia's internal Custom MeSH Identifiers (cMESHId) and the external GeneId Identifiers.

Code	Description	Data type
_id	Record Index: An incremental integer number used as a sequential index for the records in this mapping table	Integer
1	Custom MeSH ID (cMESHId): The integer identifier for the MeSH concept (often a gene, protein, or related entity) being mapped	Integer
2	NCBI Gene ID (GENEID): The unique integer identifier assigned by the NCBI Gene database for the corresponding gene	Integer

Custom MeSH to Gene Ontology Mapping Schema (cMESHId_map_GO.json)

This table provides a mapping between Quaesia's internal Custom MeSH Identifiers (cMESHId) and the external Gene Ontology (GO) Identifiers, linking biomedical subject headings to functional biological concepts.

Code	Description	Data type
_id	Record Index: An incremental integer number used as a sequential index for the records in this mapping table	Integer
1	Custom MeSH ID (cMESHId): The integer identifier for the MeSH concept being mapped (e.g., a specific biological process or molecule)	Integer
2	Gene Ontology Identifier (GO ID): The unique string ID assigned by the Gene Ontology Consortium for the corresponding functional annotation (e.g., biological process, molecular function, or cellular component)	String

Custom MeSH to Original MeSH Id Mapping Schema (cMESHId_map_MESH.json)

This table provides a mapping between Quaesia's internal Custom MeSH Identifiers (cMESHId) and the external, original MeSH Unique Identifiers (UI).

Code	Description	Data type
_id	Record Index: An incremental integer number used as a sequential index for the records in this mapping table	Integer
1	Custom MeSH ID (cMESHId): The integer identifier for the MeSH concept being mapped	Integer
2	MeSH Unique Identifier (MESH ID): The original unique identifier (UI) assigned by the National Library of Medicine (NLM) for the corresponding MeSH concept	String

Custom MeSH to PubChem CID Mapping Schema (cMESHId_map_PubChemCID.json)

This table provides a mapping between Quaesia's internal Custom MeSH Identifiers (cMESHId) and the external PubChem Compound Identifiers (CIDs), linking biomedical subject headings to chemical structure records.

Code	Description	Data type
_id	Record Index: An incremental integer number used as a sequential index for the records in this mapping table	Integer
1	Custom MeSH ID (cMESHId): The integer identifier for the MeSH concept (usually a chemical or drug) being mapped	Integer
2	PubChem Compound Identifier (CID): The unique integer ID assigned by the PubChem database for the corresponding compound	Integer

Custom MeSH to MeSH Tree Number Mapping Schema (cMESHId_map_TN.json)

This table provides a mapping between Quaesia's internal Custom MeSH Identifiers (cMESHId) and the hierarchical MeSH Tree Numbers (TN), defining the concept's position in the MeSH hierarchy.
Note on Tree Number: To maintain hierarchical structure for Supplementary Concepts (which do not have native Tree Numbers), a Tree Number is synthesized. This is achieved by taking the parent's Tree Number and concatenating it with a dot (.) and the MeSH Identifier (e.g. D05.750.716.822.111.C472990).

Code	Description	Data type
_id	Record Index: An incremental integer number used as a sequential index for the records in this mapping table	Integer
1	Custom MeSH ID (cMESHId): The integer identifier for the MeSH concept being mapped	Integer
2	MeSH Tree Number (TN): The unique string identifier that specifies the hierarchical location of the MeSH concept (e.g., C04.557.568.125)	String

Custom MeSH to Taxonomy ID Mapping Schema (cMESHId_map_TaxId.json)

This table provides a mapping between Quaesia's internal Custom MeSH Identifiers (cMESHId) and the external NCBI Taxonomy Identifiers (TaxId), linking biomedical subject headings to specific species or taxonomic groups.

Code	Description	Data type
_id	Record Index: An incremental integer number used as a sequential index for the records in this mapping table	Integer
1	Custom MeSH ID (cMESHId): The integer identifier for the MeSH concept being mapped (e.g., a specific organism, protein, or disease related to a species)	Integer
2	Taxonomy Identifier (TaxId): The unique string ID assigned by the NCBI Taxonomy database for the corresponding species or taxonomic node	String

Custom MeSH to UMLS Mapping Schema (cMESHId_map_UMLS.json)

This table provides a mapping between Quaesia's internal Custom MeSH Identifiers (cMESHId) and the external UMLS Concept Identifiers (CUIs), linking MeSH terms to the broader biomedical terminology within the Unified Medical Language System.

Code	Description	Data type
_id	Record Index: An incremental integer number used as a sequential index for the records in this mapping table	Integer
1	Custom MeSH ID (cMESHId): The integer identifier for the MeSH concept being mapped	Integer
2	UMLS Concept Identifier (CUI): The unique string identifier assigned by the UMLS Metathesaurus (often formatted as Cxxxxxxx	String

Custom MeSH to UniProt Mapping Schema (cMESHId_map_UNIPROT.json)

This table provides a mapping between Quaesia's internal Custom MeSH Identifiers (cMESHId) and the external UniProt Knowledgebase Identifiers, linking biomedical subject headings to specific protein records.

Code	Description	Data type
_id	Record Index: An incremental integer number used as a sequential index for the records in this mapping table	Integer
1	Custom MeSH ID (cMESHId): The integer identifier for the MeSH concept (often a protein or peptide) being mapped	Integer
2	UniProt Identifier: The unique string ID assigned by the UniProt Knowledgebase for the corresponding protein sequence record (e.g., a UniProt Accession number)	String

Custom MeSH to UniProt Gene Name Mapping Schema (cMESHId_map_UNIPROTGeneName.json)

This table provides a mapping between Quaesia's internal Custom MeSH Identifiers (cMESHId) and the external UniProt Preferred Gene Names.

Code	Description	Data type
_id	Record Index: An incremental integer number used as a sequential index for the records in this mapping table	Integer
1	Custom MeSH ID (cMESHId): The integer identifier for the MeSH concept (usually a gene or protein) being mapped	Integer
2	UniProt Gene Name: The preferred, official gene symbol associated with the corresponding protein record in the UniProt Knowledgebase	String

Term to Custom MeSH ID List Schema (t_is_lcMESHId.json)

This table serves as a reverse index, mapping a specific term (word or phrase) to all associated Custom MeSH Identifiers (cMESHId) found within the database. This allows for quick retrieval of all MeSH concepts related to a search term.

Code	Description	Data type
_id	Term: The unique word or phrase (e.g., a synonym or root) that serves as the primary key for the index	String
l	Custom MeSH ID List: A list of the Custom MeSH IDs (cMESHId) that are linked to the defined term	Array ([Integer...])