UPDATE: We now include Diagnostic Implications and Prognostic Implications during the annotation process
UPDATE: API token required, please see OncoKB API section for more information
For python 3
pip install -r requirements/common.txt -r requirements/pip3.txt
For python 2.7
pip install -r requirements/common.txt -r requirements/pip2.7.txt
Example input files are under data. An example script is here: example.sh
Annotates variants in MAF(https://docs.gdc.cancer.gov/Data/File_Formats/MAF_Format/) with OncoKB annotation. Supports both python2 and python3.
Get more details on the command line using python MafAnnotator.py -h
.
We recommend processing VCF files by vcf2maf with OncoKB isoforms before using the MafAnnotator
here.
You can still use MAF format to annotate atypical alterations, such as MSI-H, TMB-H, EGFR vIII. Please see more examples HERE.
We use GISTIC 2.0 format. For more information, please see https://docs.cbioportal.org/5.1-data-loading/data-loading/file-formats#discrete-copy-number-data.
Get more details on the command line using python CnaAnnotator.py -h
.
OncoKB offers to anntoate the strucutal variant. But in annotator, we only annotate the functional fusion.
The fusion format for intragenic deletion is GENE-intragenic
or GENE-GENE
.
For other fusions, please use GENEA-GENEB
or GENEA-GENEB Fusion
.
Get more details on the command line using python FusionAnnotator.py -h
.
You can comebine all annotation on sample/patient level using the clinical data annotator.
Get more details on the command line using python ClinicalDataAnnotator.py -h
.
OncoKB MafAnnotator supports annotating the alteration with HGVSp, HGVSp_Short, HGVSg or Genomic Change format. Please specify the query type with -q parameter.
The acceptable values are HGVSp_Short, HGVSp, HGVSg and Genomic_Change(case-insensitive). Please see data/example.sh for examples.
If you do not specify query type, the MafAnnotator will try to figure out the query type based on the headers.
For HGVSp_Short, the annotator takes alteration from the column HGVSp_Short or Alteration
For HGVSp, the annotator takes alteration from the column HGVSp or Alteration
For HGVSg, the annotator takes alteration from the column HGVSg or Alteration
For Genomic_Change, the annotator takes genomic change from columns Chromosome, Start_Position, End_Position, Reference_Allele, Tumor_Seq_Allele1 and Tumor_Seq_Allele2.
Annotation with Genomic_Change is relatively slow. We need to annotate the variant first with GenomeNexus(https://www.genomenexus.org/) then get annotation one by one. There is a plan to improve this method. If you are annotating a lot of data, please prioritize using other query type if applicable.
OncoKB MafAnnotator supports annotating the alteration with reference genome GRCh37 and GRCh38.
The annotator will get the reference genome from MAF file column NCBI_Build or Reference_Genome.
If there is no reference genome specified in the file, we will use the default reference genome through -r parameter.
You can specify the default reference genome using -r parameter (This is only applicable to MafAnnotator.py).
The acceptable values are GRCh37, GRCh38 (case in-sensitive).
If both values are not specified, the annotator will use OncoKB default reference genome which is GRCh37.
Introducing Simplified OncoKB Levels of Evidence:
- New Level 2, defined as “Standard care biomarker recommended by the NCCN or other expert panels predictive of response to an FDA-approved drug in this indication” (formerly Level 2A).
- Unified Level 3B, defined as “Standard care or investigational biomarker predictive of response to an FDA-approved or investigational drug in another indication” (combination of previous Levels 2B and 3B).
We have implemented these changes for 2 reasons:
- To be consistent with the Joint Consensus Recommendation by AMP, ASCO and CAP and the ESMO Scale for Clinical Actionability of molecular Targets (ESCAT)
- To reflect the clinical data that demonstrates patients with investigational predictive biomarkers for a specific tumor type based on compelling clinical evidence (currently Level 3A) are more likely to experience clinical benefit compared to patients with predictive biomarkers that are considered standard care in a different tumor type (previously Level 2B, now combined into Level 3B).
When you run MafAnnotator.py
, FusionAnnotator.py
and CnaAnnotator.py
, you need a token before accessing the OncoKB data via its web API. Please visit OncoKB Data Access Page for more information about how to register an account and get an OncoKB API token.
With the token listed under OncoKB Account Settings Page, you could use it in the following format.
python ${FILE_NAME.py} -i ${INPUT_FILE} -o ${OUTPUT_FILE} -b ${ONCOKB_API_TOKEN}
Column | Possible Values | Description |
---|---|---|
GENE_IN_ONCOKB | TRUE, FALSE | Whether the gene has been curated by the OncoKB Team |
VARIANT_IN_ONCOKB | TRUE, FALSE | Whether the variant has been curated by the OncoKB Team. Note: when a variant does not exist, it may still have annotations. |
MUTATION_EFFECT | Gain-of-function, Likely Gain-of-function, Loss-of-function, Likely Loss-of-function, Switch-of-function, Likely Switch-of-function, Neutral, Likely Neutral, Inconclusive, Unknown | The biological effect of a mutation/alteration on the protein function that gives rise to changes in the biological properties of cells expressing the mutant/altered protein compared to cells expressing the wildtype protein. |
ONCOGENIC | Oncogenic, Likely Oncogenic, Likely Neutral, Inconclusive, Unknown, Resistance | In OncoKB, “oncogenic” is defined as “referring to the ability to induce or cause cancer” as described in the second edition of The Biology of Cancer by Robert Weinberg (2014). |
LEVEL_* | Therapeutic implications | The leveled therapeutic implications |
HIGHEST_LEVEL | LEVEL_1, LEVEL_2, LEVEL_3A, LEVEL_3B, LEVEL_4, LEVEL_R1, LEVEL_R2 | The highest level of evidence for therapeutic implications |
CITATIONS | PMID, Abstract, Website Link | All citations related to a mutation/alteration |
LEVEL_Dx* | Tumor type the level of evidence is assigned to | The leveled diagnostic implications |
HIGHEST_DX_LEVEL | LEVEL_Dx1, LEVEL_Dx2, LEVEL_Dx3 | The highest level of evidence for diagnostic implications |
LEVEL_Px* | Tumor type the level of evidence is assigned to | The leveled prognostic implications |
HIGHEST_PX_LEVEL | LEVEL_Px1, LEVEL_Px2, LEVEL_Px3 | The highest level of evidence for prognostic implications |
The best way is to email [email protected] so all our team members can help.