From 9731b2c5dd61b675a0f0b90ec4829dd076580d0b Mon Sep 17 00:00:00 2001
From: spiekos <spiekos@google.com>
Date: Mon, 22 Jul 2024 20:49:58 -0700
Subject: [PATCH] Changes were made in the scripts to resolve errors in the
 data cleaning and formatting of the PharmGKB CSV+tMCF pairs. It also breaks
 out the phenotypes to distinguish that one is a MeSHQualifier and seven are
 MeSHSupplementaryConceptRecords, unlike the rest of the phenotypes which are
 MeSHDescriptors. Therefore these were separated into 3 CSV+tMCF pairs. In
 particular, the links to the enums and between entity types were fixed. This
 was done by initializing all nodes referenced and then pointing to them
 within the tMCF. Because of this any existence missing errors in the json
 reports can be ignored. The changes to the scripts, tMCF files, and
 documentation (README.md) for this import are part of GitHub PR 1056
 https://github.com/datacommonsorg/data/pull/1056

Schema Changes:
- Add CPICLevelEnum, DosageGuidelineSourceCpicNoRecommendation, DrugTypeEnum, PGxLevelEnum, PharmacogeneticAssociationEnum.
- Add properties for clinicalAnnotationCount, clinicalAnnotationCountLevel1_2, clinicalGuidelineAnnotationCount, dosageGuideline, drugHasPrescribingInfo, drugLabelAnnotationCount, drugType, fdaTopPharmacogeneticLevel, geneticVariantAnnotationCount, hasCpicDosingGuideline, hasGenomicCoordinates, hasGeneticVariantAnnotation, hasPrescribingInfo, medicalDictionaryForRegulatoryActivitiesId, metabolicPathwayCount, pharmageneticAssociation, topClinicalAnnotationLevel, topCpicLevel, topPharmacogeneticLevel, veryImportantPharmacogeneCount.
- Remove properties for fdaTopPGxLevel, mintID, nationalClinicalTrialNumber, nationalDrugCode, nationalDrugFileReferenceTerminologyCode, neuroMabID, patentID,pharmGkbClinicalAnnotationCount,pharmGkbPathwayCount, pkgbTags.

PiperOrigin-RevId: 655002332
---
 biomedical_schema/chemical_compound.mcf      | 209 ++++++++++++-----
 biomedical_schema/chemical_compound_enum.mcf | 227 +++++++++++++++----
 2 files changed, 328 insertions(+), 108 deletions(-)

diff --git a/biomedical_schema/chemical_compound.mcf b/biomedical_schema/chemical_compound.mcf
index d51b02815..dcf992c5a 100644
--- a/biomedical_schema/chemical_compound.mcf
+++ b/biomedical_schema/chemical_compound.mcf
@@ -262,6 +262,30 @@ rangeIncludes: schema:Text
 domainIncludes: dcs:ChemicalCompound
 description: "ChEMBL identifier."
 
+Node: dcid:clinicalAnnotationCount
+name: "clinicalAnnotationCount"
+typeOf: schema:Property
+domainIncludes: dcs:BiomedicalEntity
+rangeIncludes: schema:Number
+description: "Clinical annotations provide information about variant-drug pairs based on variant annotations and incorporating variant-specific prescribing guidance from clinical guidelines and FDA approved drug labels, when available. PharmGKB scientific curators manually review variant annotations and create genotype-based summaries describing the phenotypic impact of the variant. The clinical annotation is given a score, based on the scores of the supporting annotations. A clinical annotation's score is used by PharmGKB curators in the process of assigning a Level of Evidence to the annotation. Levels range from 1-4, with level 1 meeting the highest criteria. This is the number of such reports associated with a given drug or genetic variant."
+descriptionUrl: "https://www.pharmgkb.org/clinicalAnnotations"
+
+Node: dcid:clinicalAnnotationCountLevel1_2
+name: "clinicalAnnotationCountLevel1_2"
+typeOf: schema:Property
+domainIncludes: dcs:BiomedicalEntity
+rangeIncludes: schema:Number
+description: "Clinical annotations provide information about variant-drug pairs based on variant annotations and incorporating variant-specific prescribing guidance from clinical guidelines and FDA approved drug labels, when available. PharmGKB scientific curators manually review variant annotations and create genotype-based summaries describing the phenotypic impact of the variant. The clinical annotation is given a score, based on the scores of the supporting annotations. A clinical annotation's score is used by PharmGKB curators in the process of assigning a Level of Evidence to the annotation. Levels range from 1-4, with level 1 meeting the highest criteria. This is the number of such reports associated with a given drug or genetic variant whose evidence is at Level 1A, 1B, 2A, or 2B."
+descriptionUrl: "https://www.pharmgkb.org/clinicalAnnotations"
+
+Node: dcid:clinicalGuidelineAnnotationCount
+name: "clinicalGuidelineAnnotationCount"
+typeOf: schema:Property
+domainIncludes: dcs:BiomedicalEntity
+rangeIncludes: schema:Number
+description: "PGx-based drug dosing guidelines published by multiple sources. Annotations present a brief summary of the genotype-based dosing recommendations, including selected excerpts from the guidelines, and links to the source publications/documents. Tags indicate if the guideline provides dosing information, states that a drug is either indicated or contraindicated, or gives other guidance based on genotype/metabolizer phenotype. This is the number of clinical guidelines associated with a given drug, genetic variant, or phenotype."
+descriptionUrl: "https://www.pharmgkb.org/guidelineAnnotations"
+
 Node: dcid:compoundID
 name: "compoundID"
 typeOf: schema:Property
@@ -314,6 +338,13 @@ domainIncludes: schema:Drug
 description: "A dosage form is the physical form in which a drug is produced and dispensed, such as a tablet, a capsule, or an injectable."
 descriptionUrl: "https://www.fda.gov/drugs/drug-approvals-and-databases/drugsfda-glossary-terms#form"
 
+Node: dcid:dosageGuideline
+name: "dosageGuideline"
+typeOf: schema:Property
+domainIncludes: dcs:Drug
+rangeIncludes: schema:Boolean
+description: "Indicates if there is a dosage guideline for a drug."
+
 Node: dcid:dosageGuidelineSource
 name: "dosageGuidelineSource"
 typeOf: schema:Property
@@ -357,6 +388,13 @@ domainIncludes: schema:Drug
 rangeIncludes: schema:Text
 description: "The name of the generic version of a drug, which is a medication with the exact same active ingredient as the brand-name drug."
 
+Node: dcid:drugHasPrescribingInfo
+name: "drugHasPrescribingInfo"
+typeOf: schema:Property
+domainIncludes: dcs:Drug
+rangeIncludes: schema:Boolean
+description: "Indicates if there is a prescribing information for a drug."
+
 Node: dcid:drugHasRxAnnotation
 name: "drugHasRxAnnotation"
 typeOf: schema:Property
@@ -365,6 +403,13 @@ rangeIncludes: schema:Boolean
 description: "The drug has a drug label containing pharmacogenetic information approved by the U.S. Food and Drug Administration (FDA), European Medicines Agencey (EMA), Swiss Agenecy of Therapeutic Products (Swissmedic), Pharmaceuticals and Medical Devices Agency, Japan (PMDA) and/or Health Canada (Sante Canada) (HCSC)."
 descriptionUrl: "https://www.pharmgkb.org/labelAnnotations"
 
+Node: dcid:drugLabelAnnotationCount
+name: "drugLabelAnnotationCount"
+typeOf: schema:Property
+domainIncludes: dcs:BiomedicalEntity
+rangeIncludes: schema:Number
+description: "PharmGKB annotates drug labels approved by the US Food and Drug Administration (FDA) containing pharmacogenetic information. Read more about PharmGKB label annotations, PGx Levels and the "tags" found in the table below. If a specific genetic allele is discussed on the label, it is listed in the "Alleles" column. PharmGKB includes the FDA "PGx Association" group, based on the FDA Table of Pharmacogenetic Associations. This is the number of such labels with which a given drug or genetic variant is associated."
+
 Node: dcid:drugLabelHasDosingInformation
 name: "drugLabelHasDosingInformation"
 typeOf: schema:Property
@@ -395,6 +440,13 @@ rangeIncludes: schema:Text
 description: "A standard term in the pharmaceutical industry for a brand name or trademark name of a drug."
 descriptionUrl: "https://en.wikipedia.org/wiki/Drug_nomenclature#Drug_brands"
 
+Node: dcid:drugType
+name: "drugType"
+typeOf: schema:Property
+domainIncludes: dcs:ChemicalCompound
+rangeIncludes: dcs:DrugTypeEnum
+description: "The therapeuatic modality of a drug."
+
 Node: dcid:electronMicroscopyDataBankID
 typeOf: schema:Property
 name: "electronMicroscopyDataBankID"
@@ -433,13 +485,13 @@ rangeIncludes: schema:Text
 domainIncludes: schema:DrugStrength
 description: "Product ID according to the FDA."
 
-Node: dcid:fdaTopPGxLevel
-name: "fdaTopPGxLevel"
+Node: dcid:fdaTopPharmacogeneticLevel
+name: "fdaTopPharmacogeneticLevel"
 typeOf: schema:Property
-domainIncludes: schema:Drug
-rangeIncludes: dcs:PharmGkbPGxLevelEnum
-description: "The 'PGx Level' tag ('Testing required', 'Testing Recommended', 'Actionable PGx' and 'Informative PGx') is the PharmGKB interpretation of the level of action implied in each drug label. In this case, this is the highest level of confidence interperted for any drug label from the U.S. Food and Drug Administration (FDA)."
-descriptionUrl: "https://www.pharmgkb.org/labelAnnotations"
+domainIncludes: dcs:Drug
+rangeIncludes: dcs:PGxLevelEnum
+description: "The 'PGx Level' tag ('Testing required', 'Testing Recommended', 'Actionable PGx' and 'Informative PGx') indicates the level of action implied in each label. This is the highest level reported by the Food and Drug Administration (FDA)."
+descriptionUrl: "https://www.pharmgkb.org/page/drugLabelLegend"
 
 Node: dcid:finalReconstitutedSolutionVolume
 name: "finalReconstitutedSolutionVolume"
@@ -448,6 +500,14 @@ rangeIncludes: schema:Quantity
 domainIncludes: schema:DrugStrength
 description: "Final reconstituted solution volume for drug dose."
 
+Node: dcid:geneticVariantAnnotationCount
+name: "geneticVariantAnnotationCount"
+typeOf: schema:Property
+domainIncludes: dcs:BiomedicalEntity
+rangeIncludes: schema:Number
+description: "Genetic Variant annotations report the association between a variant (e.g. SNP, indel, repeat, haplotype) and a drug phenotype from a single publication. Annotations are created manually by scientific curators who read each paper, extract the key information (including relevant drugs, study size, population data, statistical values, etc.), and map the variants to a common standard (typically dbSNP rsID if it exists). This is the number of such reports associated with a given drug or genetic variant."
+descriptionUrl: "https://www.pharmgkb.org/variantAnnotations"
+
 Node: dcid:geneSynonym
 name: "geneSynonym"
 typeOf: schema:Property
@@ -478,6 +538,28 @@ rangeIncludes: dcs:ActiveIngredientAmount
 domainIncludes: schema:DrugStrength
 description: "The quantity of active ingredient in the drug."
 
+Node: dcid:hasCpicDosingGuideline
+name: "hasCpicDosingGuideline"
+typeOf: schema:Property
+domainIncludes: dcs:Gene
+rangeIncludes: schema:Boolean
+description: "The Clinical Pharmacogenetics Implementation Consortium (CPIC) was established in 2009 as a shared project between PharmGKB and the Pharmacogenomics Research Network (PGRN). CPIC is funded by the NIH/NHGRI. This indicates whether a gene has a drug dosing guideline issued by the CPIC that is associated with it."
+descriptionUrl: "https://www.pharmgkb.org/page/cpic"
+
+Node: dcid:hasGenomicCoordinates
+name: "hasGenomicCoordinates"
+typeOf: schema:Property
+domainIncludes: dcs:GenomeAnnotation
+rangeIncludes: dcs:GenomicCoordinates
+description: "Genomic coordinates specify the location of the position of an element within a specified genome assembly. It is a specified set of chromosome start_position end_position."
+
+Node: dcid:hasGeneticVariantAnnotation
+name: "hasGeneticVariantAnnotation"
+typeOf: schema:Property
+domainIncludes: dcs:Gene
+rangeIncludes: schema:Boolean
+description: "Indicates whether there are gene has genetic variants that are associated with it."
+
 Node: dcid:humanCellType
 typeOf: schema:Property
 name: "humanCellType"
@@ -628,6 +710,22 @@ rangeIncludes: schema:Number
 domainIncludes: dcs:ChemicalCompound
 description: "The maximum FDA clinical trials phase in which the compound has participated for any disease."
 
+Node: dcid:medicalDictionaryForRegulatoryActivitiesId
+name: "medicalDictionaryForRegulatoryActivitiesId"
+typeOf: schema:Property
+abbreviation: "medraId"
+domainIncludes: dcs:ChemicalCompound
+rangeIncludes: schema:Text
+description: "The identifier for a chemical compound within the Medical Dictionary for Regulatory Activities (MedDRA), which is an internationally used set of terms relating to medical conditions, medicines and medical devices."
+
+Node: dcid:metabolicPathwayCount
+name: "metabolicPathwayCount"
+typeOf: schema:Property
+domainIncludes: dcs:ChemicalCompound
+rangeIncludes: schema:Number
+description: "The number of PharmGKB pathways associated with a given drug. PharmGKB pathways are evidence-based diagrams depicting the pharmacokinetics (PK) and/or pharmacodynamics (PD) of a drug with relevant (or potential) pharmacogenetic (PGx) associations. Drugs featured in PharmGKB pathways are chosen through extensive review of a variety of sources, including, but not limited to, the U.S. Food and Drug Administration (FDA) biomarker list and Clinical Pharmacogenetics Implementation Consortium (CPIC) nominations."
+descriptionUrl: "https://www.pharmgkb.org/pathways"
+
 Node: dcid:mintID
 typeOf: schema:Property
 name: "mintID"
@@ -676,21 +774,12 @@ rangeIncludes: schema:Text
 domainIncludes: dcs:Antibody
 descriptionUrl:"https://patentscope.wipo.int/search/en/search.jsf"
 
-Node: dcid:pharmGkbClinicalAnnotationCount
-name: "pharmGkbClinicalAnnotationCount"
-typeOf: schema:Property
-domainIncludes: schema:Drug
-rangeIncludes: schema:Number
-description: "The number of clinical annotations associated with a drug as reported by PharmGKB. PharmGKB clinical annotations provide information about variant-drug pairs based on variant annotations and incorporating variant-specific prescribing guidance from clinical guidelines and FDA approved drug labels, when available. PharmGKB scientific curators manually review variant annotations and create genotype-based summaries describing the phenotypic impact of the variant."
-descriptionUrl: "https://www.pharmgkb.org/clinicalAnnotations"
-
-Node: dcid:pharmGkbPathwayCount
-name: "pharmGkbPathwayCount"
+Node: dcid:pharmacogeneticAssociation
+name: "pharmacogeneticAssociation"
 typeOf: schema:Property
-domainIncludes: schema:Drug
-rangeIncludes: schema:Number
-description: "The number of PharmGKB pathways associated with a given drug. PharmGKB pathways are evidence-based diagrams depicting the pharmacokinetics (PK) and/or pharmacodynamics (PD) of a drug with relevant (or potential) pharmacogenetic (PGx) associations. Drugs featured in PharmGKB pathways are chosen through extensive review of a variety of sources, including, but not limited to, the U.S. Food and Drug Administration (FDA) biomarker list and Clinical Pharmacogenetics Implementation Consortium (CPIC) nominations."
-descriptionUrl: "https://www.pharmgkb.org/pathways"
+domainIncludes: dcs:Drug
+rangeIncludes: dcs:PharmacogeneticAssociationEnum
+description: "These associaitons are based off of the FDA Table of Pharmacogenetic Associations."
 
 Node: dcid:pharmGkbRareVariantScoringRange
 name: "pharmGkbRareVariantScoringRange"
@@ -700,45 +789,6 @@ rangeIncludes: schema:Text
 description: "The assignment of clinical annotation levels of evidence (LOE) is primarily informed by the PharmGKB annotation scoring system for clinical annotations and variant annotations. This reports the scoring range that determines the level for rare genetic variants defined as having a minor allele frequenct (MAF) <0.01 or <1%."
 descriptionUrl: "https://www.pharmgkb.org/page/clinAnnLevels"
 
-Node: dcid:pharmGkbTopPGxLevel
-name: "pharmGkbTopPGxLevel"
-typeOf: schema:Property
-domainIncludes: schema:Drug
-rangeIncludes: dcs:PharmGkbPGxLevelEnum
-description: "The 'PGx Level' tag ('Testing required', 'Testing Recommended', 'Actionable PGx' and 'Informative PGx') is the PharmGKB interpretation of the level of action implied in each drug label. In this case, this is the highest level of confidence interperted for any drug label from any source."
-descriptionUrl: "https://www.pharmgkb.org/labelAnnotations"
-
-Node: dcid:pharmGkbTopClinicalAnnotationLevel
-name: "pharmGkbTopClinicalAnnotationLevel"
-typeOf: schema:Property
-domainIncludes: schema:Drug
-rangeIncludes: dcs:PharmGkbClinicalLevelEnum
-description: "PharmGKB clinical annotations provide information about variant-drug pairs based on variant annotations and incorporating variant-specific prescribing guidance from clinical guidelines and FDA approved drug labels, when available. PharmGKB scientific curators manually review variant annotations and create genotype-based summaries describing the phenotypic impact of the variant. The clinical annotation is given a score, based on the scores of the supporting annotations. A clinical annotation's score is used by PharmGKB curators in the process of assigning a Level of Evidence to the annotation. Levels range from 1-4, with level 1 meeting the highest criteria. The highest level of clinical annotation of any variant associated with a drug is reported here."
-descriptionUrl: "https://www.pharmgkb.org/clinicalAnnotations"
-
-Node: dcid:pharmGkbVariantAnnotationCount
-name: "pharmGkbVariantAnnotationCount"
-typeOf: schema:Property
-domainIncludes: schema:Drug
-rangeIncludes: schema:Number
-description: "The number of variants with a given drug as reported by PharmGKB. PharmGKB clinical annotations provide information about variant-drug pairs based on variant annotations and incorporating variant-specific prescribing guidance from clinical guidelines and FDA approved drug labels, when available. PharmGKB scientific curators manually review variant annotations and create genotype-based summaries describing the phenotypic impact of the variant."
-descriptionUrl: "https://www.pharmgkb.org/variantAnnotations"
-
-Node: dcid:pharmGkbVeryImportantPharmacogeneCount
-name: "pharmGkbVeryImportantPharmacogeneCount"
-typeOf: schema:Property
-domainIncludes: schema:Drug
-rangeIncludes: schema:Number
-description: "The number of Very Important Pharmacogene (VIP) summaries assocated with a given drug as reported by PharmGKB. VIP summaries provide an overview of a significant gene involved in metabolism of, or response to, one or several drugs. Often, VIPs either play a role in the metabolism of many drugs (e.g. CYP2D6), or contain variants which potentially contribute to a severe drug response (e.g. HLA-B). VIP summaries typically include background information on the gene including any disease associations, as well as in-depth information on the gene's pharmacogenetics. Many VIP summaries have been published in the journal Pharmacogenetics and Genomics."
-descriptionUrl: "https://www.pharmgkb.org/vips"
-
-Node: dcid:pkgbTags
-name: "pkgbTags"
-typeOf: schema:Property
-domainIncludes: dcs:ChemicalCompound
-rangeIncludes: schema:Text
-description: "A tag associated with a chemical compound, which is used internally by the database PharmGKB."
-
 Node: dcid:prositeID
 typeOf: schema:Property
 name: "prositeID"
@@ -847,6 +897,15 @@ domainIncludes: dcs:InteractionDetectionMethodEnum,dcs:InteractionTypeEnum,dcs:I
 description: "Identifiers of the RESID database which contains the collections of annotations and structures for protein modifications."
 descriptionUrl: "https://proteininformationresource.org/resid/"
 
+Node: dcid:rxConceptUniqueId
+name: "rxConceptUniqueId"
+typeOf: schema:Property
+abbreviation: "rxCUI"
+domainIncludes: dcs:ChemicalCompound
+rangeIncludes: schema:Text
+description: "RxNorm is two things: a normalized naming system for generic and branded drugs; and a tool for supporting semantic interoperation between drug terminologies and pharmacy knowledge base systems. The National Library of Medicine (NLM) produces RxNorm. This is the Concept Unique Identifier for a concept in RxNorm."
+descriptionUrl: "https://www.nlm.nih.gov/research/umls/rxnorm/overview.html"
+
 Node: dcid:rxNormId
 name: "rxNormId"
 typeOf: schema:Property
@@ -891,6 +950,30 @@ domainIncludes: schema:DrugStrength
 description: "The therapeutic equivalence code indicates whether the FDA has evaluated a particular approved product (e.g., a particular strength of an approved drug) as therapeutically equivalent to other pharmaceutically equivalent products and to provide additional information on the basis of FDA's evaluations (second letter)."
 descriptionUrl: "https://www.fda.gov/drugs/development-approval-process-drugs/orange-book-preface#TEC"
 
+Node: dcid:topClinicalAnnotationLevel
+name: "topClinicalAnnotationLevel"
+typeOf: schema:Property
+domainIncludes: dcs:ChemicalCompound
+rangeIncludes: dcs:PharmGkbClinicalLevelEnum
+description: "PharmGKB clinical annotations provide information about variant-drug pairs based on variant annotations and incorporating variant-specific prescribing guidance from clinical guidelines and FDA approved drug labels, when available. PharmGKB scientific curators manually review variant annotations and create genotype-based summaries describing the phenotypic impact of the variant. The clinical annotation is given a score, based on the scores of the supporting annotations. A clinical annotation's score is used by PharmGKB curators in the process of assigning a Level of Evidence to the annotation. Levels range from 1-4, with level 1 meeting the highest criteria. The highest level of clinical annotation of any variant associated with a drug is reported here."
+descriptionUrl: "https://www.pharmgkb.org/clinicalAnnotations"
+
+Node: dcid:topCpicLevel
+name: "topCpicLevel"
+typeOf: schema:Property
+domainIncludes: dcs:Drug
+rangeIncludes: dcs:CPICLevelEnum
+description: "The Clinical Pharmacogenetics Implementation Consortium (CPIC) assigns CPIC levels to gene/drug pairs. The levels (A, B, C, and D) assigned are subject to change; only those gene/drug pairs that have been the subject of guidelines have had sufficient in-depth review of evidence to provide definitive CPIC level assignments. Note that only CPIC level A and B gene/drug pairs have sufficient evidence for at least one prescribing action to be recommended. CPIC level C and D gene/drug pairs are not considered to have adequate evidence or actionability to have prescribing recommendations."
+descriptionUrl: "https://cpicpgx.org/prioritization/"
+
+Node: dcid:topPharmacogeneticLevel
+name: "topPharmacogeneticLevel"
+typeOf: schema:Property
+domainIncludes: dcs:Drug
+rangeIncludes: dcs:PGxLevelEnum
+description: "The 'PGx Level' tag ('Testing required', 'Testing Recommended', 'Actionable PGx' and 'Informative PGx') indicates the level of action implied in each label. This is the highest level reported by any reporting agency."
+descriptionUrl: "https://www.pharmgkb.org/page/drugLabelLegend"
+
 Node: dcid:uniProtID
 name: "uniProtID"
 typeOf: schema:Property
@@ -912,6 +995,14 @@ domainIncludes: dcs:ChemicalCompoundGeneticVariantAssociation,dcs:GeneGeneticVar
 rangeIncludes: dcs:GeneticVariant
 description: "Link to a GeneticVariant node for a genetic variant that is associated with another entity."
 
+Node: dcid:veryImportantPharmacogeneCount
+name: "veryImportantPharmacogeneCount"
+typeOf: schema:Property
+domainIncludes: dcs:ChemicalCompound
+rangeIncludes: schema:Number
+description: "The number of Very Important Pharmacogene (VIP) summaries assocated with a given drug as reported by PharmGKB. VIP summaries provide an overview of a significant gene involved in metabolism of, or response to, one or several drugs. Often, VIPs either play a role in the metabolism of many drugs (e.g. CYP2D6), or contain variants which potentially contribute to a severe drug response (e.g. HLA-B). VIP summaries typically include background information on the gene including any disease associations, as well as in-depth information on the gene's pharmacogenetics. Many VIP summaries have been published in the journal Pharmacogenetics and Genomics."
+descriptionUrl: "https://www.pharmgkb.org/vips"
+
 Node: dcid:wordElementType
 name: "wordElementType"
 typeOf: schema:Property
diff --git a/biomedical_schema/chemical_compound_enum.mcf b/biomedical_schema/chemical_compound_enum.mcf
index 3d1eaa8d9..0ddf732e2 100644
--- a/biomedical_schema/chemical_compound_enum.mcf
+++ b/biomedical_schema/chemical_compound_enum.mcf
@@ -1222,6 +1222,65 @@ name: "IntAct MiScore"
 description: "The confidence score of a protein-protein interaction given in IntAct Molecular Interaction Database."
 
 
+
+#CPICLevelEnum
+Node: dcid:CPICLevelEnum
+name: "CPICLevelEnum"
+typeOf: schema:Class
+subClassOf: schema:Enumeration
+description: "The Clinical Pharmacogenetics Implementation Consortium (CPIC) assigns CPIC levels to gene/drug pairs. The levels (A, B, C, and D) assigned are subject to change; only those gene/drug pairs that have been the subject of guidelines have had sufficient in-depth review of evidence to provide definitive CPIC level assignments. Note that only CPIC level A and B gene/drug pairs have sufficient evidence for at least one prescribing action to be recommended. CPIC level C and D gene/drug pairs are not considered to have adequate evidence or actionability to have prescribing recommendations."
+
+Node: dcid:CPICLevelA
+name: "CPIC Level A"
+typeOf: dcs:CPICLevelEnum
+abbreviation: "A"
+description: "Genetic information should be used to change prescribing of affected drug. The preponderance of evidence is high or moderate in favor of changing prescribing. At least one moderate or strong action (change in prescribing) recommended."
+descriptionUrl: "https://cpicpgx.org/prioritization/"
+
+Node: dcid:CPICLevelA_B
+name: "CPIC Level A/B"
+typeOf: dcs:CPICLevelEnum
+abbreviation: "A/B"
+description: "Preliminary review indicates it is likely that the definitive CPIC level will be either A or B. Full evidence review needed to assess level of evidence, but prescribing actionability is likely. Full review by expert guideline group to assign strength of recommendation."
+descriptionUrl: "https://cpicpgx.org/prioritization/"
+
+Node: dcid:CPICLevelB
+name: "CPIC Level B"
+typeOf: dcs:CPICLevelEnum
+abbreviation: "B"
+description: "Genetic information could be used to change prescribing of the affected drug because alternative therapies/dosing are extremely likely to be as effective and as safe as non-genetically based dosing. The preponderance of evidence is weak with little conflicting data. At least one optional action (change in prescribing) is recommended."
+descriptionUrl: "https://cpicpgx.org/prioritization/"
+
+Node: dcid:CPICLevelB_C
+name: "CPIC Level B/C"
+typeOf: dcs:CPICLevelEnum
+abbreviation: "B/C"
+description: "Preliminary review indicates it is likely that the definitive CPIC level will be either B or C. Prescribing actionability based on genetics is not clear without further evidence review. Full review by expert guideline group to assess strength of recommendation."
+descriptionUrl: "https://cpicpgx.org/prioritization/"
+
+Node: dcid:CPICLevelC
+name: "CPIC Level C"
+typeOf: dcs:CPICLevelEnum
+abbreviation: "C"
+description: "There are published studies at varying levels of evidence, some with mechanistic rationale, but no prescribing actions are recommended because (a) dosing based on genetics makes no convincing difference or (b) alternatives are unclear, possibly less effective, more toxic, or otherwise impractical or (c) few published studies or mostly weak evidence and clinical actions are unclear. Most important for genes that are subject of other CPIC guidelines or genes that are commonly included in clinical or DTC tests. Evidence levels can vary. No prescribing actions are recommended."
+descriptionUrl: "https://cpicpgx.org/prioritization/"
+
+Node: dcid:CPICLevelC_D
+name: "CPIC Level C/D"
+typeOf: dcs:CPICLevelEnum
+abbreviation: "C/D"
+description: "Preliminary review indicates it is likely that the definitive CPIC level will be either C or D. Evidence levels can vary. No prescribing actions are recommended"
+descriptionUrl: "https://cpicpgx.org/prioritization/"
+
+Node: dcid:CPICLevelD
+name: "CPIC Level D"
+typeOf: dcs:CPICLevelEnum
+abbreviation: "D"
+description: "There are few published studies, clinical actions are unclear, little mechanistic basis, mostly weak evidence, or substantial conflicting data. If the genes are not widely tested for clinically, evaluations are not needed. Evidence levels can vary. No prescribing actions are recommended."
+descriptionUrl: "https://cpicpgx.org/prioritization/"
+
+
+
 # DosageFormEnum
 Node: dcid:DosageFormEnum
 typeOf: schema:Class
@@ -2250,6 +2309,57 @@ name: "Spanish Pharmacogenetics and Pharmacogenomics Society (SEFF) and the Span
 typeOf: dcs:DosageGuidelineSourceEnum
 abbreviation: "SEFF/SEOM"
 
+# DosageGuidelineSourceEnum
+Node: dcid:DosageGuidelineSourceCpicNoRecommendation
+name: "CPIC No Recommendation"
+typeOf: dcs:DosageGuidelineSourceEnum
+description: "There is insufficient evidence, confidence, or agreement to provide a recommendation to guide clinical practice at this time."
+descriptionUrl: "https://cpicpgx.org/strength-of-recommendations"
+
+
+
+# DrugTypeEnum
+Node: dcid:DrugTypeEnum
+name: "DrugTypeEnum"
+typeOf: schema:Class
+subClassOf: schema:Enumeration
+description: "The therapeuatic modality of a drug."
+
+Node: dcid:DrugTypeBiologicalIntermediate
+name: "Biological Intermediate"
+typeOf: dcs:DrugTypeEnum
+description: "A material produced during a manufacturing process that is not the drug substance or the drug product but for which manufacture is critical to the successful production of the drug substance or the drug product."
+descriptionUrl: "https://en.ennov.com/glossary/intermediate-or-intermediate-product/"
+
+Node: dcid:DrugTypeIon
+name: "Ion"
+typeOf: dcs:DrugTypeEnum
+description: "An ion is an atom or molecule with a net electrical charge."
+descriptionUrl: "https://en.wikipedia.org/wiki/Ion"
+
+Node: dcid:DrugTypeMetabolite
+name: "Metabolite"
+typeOf: dcs:DrugTypeEnum
+description: "A metabolite is an intermediate or end product of metabolism. The term is usually used for small molecules. Metabolites have various functions, including fuel, structure, signaling, stimulatory and inhibitory effects on enzymes, catalytic activity of their own (usually as a cofactor to an enzyme), defense, and interactions with other organisms (e.g. pigments, odorants, and pheromones).."
+descriptionUrl: "https://en.wikipedia.org/wiki/Metabolite"
+
+Node: dcid:DrugTypeProdrug
+name: "Prodrug"
+typeOf: dcs:DrugTypeEnum
+description: "A prodrug is a pharmacologically inactive medication or compound that, after intake, is metabolized (i.e., converted within the body) into a pharmacologically active drug. Instead of administering a drug directly, a corresponding prodrug can be used to improve how the drug is absorbed, distributed, metabolized, and excreted (ADME)."
+descriptionUrl: "https://en.wikipedia.org/wiki/Prodrug"
+
+Node: dcid:DrugTypeSmallMolecule
+name: "Small Molecule"
+typeOf: dcs:DrugTypeEnum
+description: "A drug that can enter cells easily because it has a low molecular weight. Once inside the cells, it can affect other molecules, such as proteins, and may cause cancer cells to die. This is different from drugs that have a large molecular weight, which keeps them from getting inside cells easily. Many targeted therapies are small-molecule drugs."
+descriptionUrl: "https://www.cancer.gov/publications/dictionaries/cancer-terms/def/small-molecule-drug"
+
+Node: dcid:DrugTypeUnknown
+name: "Unknown"
+typeOf: dcs:DrugTypeEnum
+
+
 
 # the instances of HumanCellTypeEnum are in HumanCellTypeEnum.mcf
 Node: dcid:HumanCellTypeEnum
@@ -2356,35 +2466,31 @@ name: "None (Tentative Approval)"
 description: "Drug products that have been tentatively approved."
 descriptionUrl: "https://www.fda.gov/drugs/drug-approvals-and-databases/drugsfda-glossary-terms#M"
 
-Node: dcid:ProteinExpressionScoreEnum
+
+
+# PGxLevelEnum
+Node: dcid:PGxLevelEnum
+name: "PGxLevelEnum"
 typeOf: schema:Class
 subClassOf: schema:Enumeration
-name: "ProteinExpressionScoreEnum"
-description: "The level of protein observed in a specified cell type."
+description: "The 'PGx Level' tag ('Testing required', 'Testing Recommended', 'Actionable PGx' and 'Informative PGx') indicates the level of action implied in each label."
 
-Node: dcid:ProteinExpressionNotDetected
-typeOf: dcs:ProteinExpressionScoreEnum
-name: "Not Detected"
-description: "The protein expression score is ‘Not detected' when the combination of intensity and fractions is negative or weak and <25%. Protein expression score is based on immunohistochemical data manually scored with regard to staining intensity (negative, weak, moderate or strong) and fraction of stained cells (<25%, 25-75% or >75%)."
-descriptionUrl: "https://www.proteinatlas.org/about/help#4"
+Node: dcid:PGxLevelTestingRequired
+name: "Testing Required"
+typeOf: dcs:PGxLevelEnum
 
-Node: dcid:ProteinExpressionLow
-typeOf: dcs:ProteinExpressionScoreEnum
-name: "Low"
-description: "The protein expression score is ‘Low' when the combination of intensity and fractions is weak combined with either 25 - 75% or 75%, or moderate <25%. Protein expression score is based on immunohistochemical data manually scored with regard to staining intensity (negative, weak, moderate or strong) and fraction of stained cells (<25%, 25-75% or >75%)"
-descriptionUrl: "https://www.proteinatlas.org/about/help#4"
+Node: dcid:PGxLevelTestingRecommended
+name: "Testing Recommended"
+typeOf: dcs:PGxLevelEnum
 
-Node: dcid:ProteinExpressionMedium
-typeOf: dcs:ProteinExpressionScoreEnum
-name: "Medium"
-description: "The protein expression score is ‘Medium' when the combination of intensity and fractions is moderate combined with either 25 - 75% or 75% - medium, or strong <25%. Protein expression score is based on immunohistochemical data manually scored with regard to staining intensity (negative, weak, moderate or strong) and fraction of stained cells (<25%, 25-75% or >75%)"
-descriptionUrl: "https://www.proteinatlas.org/about/help#4"
+Node: dcid:PGxLevelActionablePGx
+name: "Actionable PGx"
+typeOf: dcs:PGxLevelEnum
+
+Node: dcid:PGxLevelInformativePGx
+name: "Informative PGx"
+typeOf: dcs:PGxLevelEnum
 
-Node: dcid:ProteinExpressionHigh
-typeOf: dcs:ProteinExpressionScoreEnum
-name: "High"
-description: "The protein expression score is ‘Medium' when the combination of intensity and fractions is strong combined with either 25 - 75% or 75%. Protein expression score is based on immunohistochemical data manually scored with regard to staining intensity (negative, weak, moderate or strong) and fraction of stained cells (<25%, 25-75% or >75%)"
-descriptionUrl: "https://www.proteinatlas.org/about/help#4"
 
 
 # PharmGkbClinicalLevelEnum
@@ -2443,36 +2549,59 @@ pharmGkbStandardScoringRange: "<0"
 pharmGkbRareVariantScoringRange: "<0"
 
 
-# PharmGkbPGxLevelEnum
-Node: dcid:PharmGkbPGxLevelEnum
-name: "PharmGkbPGxLevelEnum"
-typeOf: schema:Enumeration
-description: "The 'PGx Level' tag ('Testing required', 'Testing Recommended', 'Actionable PGx' and 'Informative PGx') is the PharmGKB interpretation of the level of action implied in each drug label."
-descriptionUrl: "https://www.pharmgkb.org/labelAnnotations"
 
-Node: dcid:PharmGkbPGxLevelTestingRequired
-name: "Testing Required"
-typeOf: dcs:PharmGkbPGxLevelEnum
-description: "The label states or implies that some sort of gene, protein or chromosomal testing, including genetic testing, functional protein assays, cytogenetic studies, etc., should be conducted on the gene or gene product mentioned in the annotation before using this drug. This requirement may only be for a particular subset of patients. If a label describes the presence or absence of a gene, genetic variant or gene product as an indication for the drug, PharmGKB considers the label to imply a testing requirement. If the label states a test "should be" performed, this is also interpreted as a requirement."
-descriptionUrl: "https://www.pharmgkb.org/page/drugLabelLegend#pgx-level"
+# PharmacogeneticAssociationEnum
+Node: dcid:PharmacogeneticAssociationEnum
+name: "PharmacogeneticAssociationEnum"
+typeOf: schema:Class
+subClassOf: schema:Enumeration
+description: "These associaitons are based off of the FDA Table of Pharmacogenetic Associations."
 
-Node: dcid:PharmGkbPGxLevelTestingRecommended
-name: "Testing Recommended"
-typeOf: dcs:PharmGkbPGxLevelEnum
-description: "The label states or implies that some sort of gene, protein or chromosomal testing, including genetic testing, functional protein assays, cytogenetic studies, etc., is recommended to be carried out on the gene or gene product mentioned in the annotation before using this drug. This recommendation may only be for a particular subset of patients. PharmGKB considers labels that say testing "should be considered" or "Consider genotyping or phenotyping" to be recommending testing."
-descriptionUrl: "https://www.pharmgkb.org/page/drugLabelLegend#pgx-level"
+Node: dcid:PharmacogeneticAssociationParmacokineticOnly
+name: "Parmacokinetic Only"
+typeOf: dcs:PharmacogeneticAssociationEnum
 
-Node: dcid:PharmGkbPGxLevelActionablePGx
-name: "Actionable PGx"
-typeOf: dcs:PharmGkbPGxLevelEnum
-description: "The label may contain information about changes in efficacy, dosage, metabolism or toxicity due to gene/protein/chromosomal variants or phenotypes (e.g. "poor metabolizers"). Or the label may mention contraindication of the drug in a particular subset of patients with particular variants/genotypes/phenotypes. However, the label does not require or recommend gene, protein or chromosomal testing."
-descriptionUrl: "https://www.pharmgkb.org/page/drugLabelLegend#pgx-level"
+Node: dcid:PharmacogeneticAssociationPotentialSafetyImpact
+name: "Potential Safety Impact"
+typeOf: dcs:PharmacogeneticAssociationEnum
+
+Node: dcid:PharmacogeneticAssociationTherapeuticManagementRecommended
+name: "Therapeutic Management Recommended"
+typeOf: dcs:PharmacogeneticAssociationEnum
+
+
+
+# ProteinExpressionScoreEnum
+Node: dcid:ProteinExpressionScoreEnum
+typeOf: schema:Class
+subClassOf: schema:Enumeration
+name: "ProteinExpressionScoreEnum"
+description: "The level of protein observed in a specified cell type."
+
+Node: dcid:ProteinExpressionNotDetected
+typeOf: dcs:ProteinExpressionScoreEnum
+name: "Not Detected"
+description: "The protein expression score is ‘Not detected' when the combination of intensity and fractions is negative or weak and <25%. Protein expression score is based on immunohistochemical data manually scored with regard to staining intensity (negative, weak, moderate or strong) and fraction of stained cells (<25%, 25-75% or >75%)."
+descriptionUrl: "https://www.proteinatlas.org/about/help#4"
+
+Node: dcid:ProteinExpressionLow
+typeOf: dcs:ProteinExpressionScoreEnum
+name: "Low"
+description: "The protein expression score is ‘Low' when the combination of intensity and fractions is weak combined with either 25 - 75% or 75%, or moderate <25%. Protein expression score is based on immunohistochemical data manually scored with regard to staining intensity (negative, weak, moderate or strong) and fraction of stained cells (<25%, 25-75% or >75%)"
+descriptionUrl: "https://www.proteinatlas.org/about/help#4"
+
+Node: dcid:ProteinExpressionMedium
+typeOf: dcs:ProteinExpressionScoreEnum
+name: "Medium"
+description: "The protein expression score is ‘Medium' when the combination of intensity and fractions is moderate combined with either 25 - 75% or 75% - medium, or strong <25%. Protein expression score is based on immunohistochemical data manually scored with regard to staining intensity (negative, weak, moderate or strong) and fraction of stained cells (<25%, 25-75% or >75%)"
+descriptionUrl: "https://www.proteinatlas.org/about/help#4"
+
+Node: dcid:ProteinExpressionHigh
+typeOf: dcs:ProteinExpressionScoreEnum
+name: "High"
+description: "The protein expression score is ‘Medium' when the combination of intensity and fractions is strong combined with either 25 - 75% or 75%. Protein expression score is based on immunohistochemical data manually scored with regard to staining intensity (negative, weak, moderate or strong) and fraction of stained cells (<25%, 25-75% or >75%)"
+descriptionUrl: "https://www.proteinatlas.org/about/help#4"
 
-Node: dcid:PharmGkbPGxLevelInformativePGx
-name: "Informative PGx"
-typeOf: dcs:PharmGkbPGxLevelEnum
-description: "The drug label meets one of two criteria: The label contains information stating that particular gene/protein/chromosomal variants or metabolizer phenotypes do not affect a drug's efficacy, dosage, metabolism or toxicity. Or, the label states that particular variants or phenotypes affect a drug's efficacy, dosage, metabolism or toxicity, but this effect is not “clinically” significant. OR The label appears or appeared on the FDA Biomarker List but does not currently meet the requirements to be assigned as “Testing required”, “Testing recommended” or “Actionable PGx”. PharmGKB annotates every label that appears on the FDA Biomarker list, regardless of whether we would otherwise annotate the label."
-descriptionUrl: "https://www.pharmgkb.org/page/drugLabelLegend#pgx-level"
 
 
 # ProteinTypeEnum