Tool for automated classification of genetic variants according to ACMG criteria (implemented with open-cravat). This tool takes VCF file as input, annotates variants with several modules, assigns ACMG criteria to them and returns annotated file in TSV format.
I am currently working on deploying Diablo ACMG as conda package.
For now, the easiest way to install Diablo annotate is to git clone repository and install dependencies from .yml file.
git clone https://github.com/TheSergeyPixel/Diablo_ACMG.git
conda env create -f /path/to/cloned/repo/diablo_annotate.yml
Make sure, that you've installed all the requirements for open-cravat (check Required modules section)
Basic usage:
python Diablo_annotate.py -i /path/to/your/vcf.gz -o /output/file/name -d /path/to/db
You can also test your installation via running Diablo ACMG on test sample from core directory:
python Diablo_annotate.py -i test/IDTEPF.vcf.gz -o test/test.tsv
It's important to note, that we don't support multi-sample VCFs yet.
Description of options:
| Option | Description |
|---|---|
-i |
Path to your input VCF file or TSV annotated file (latter case assumes, that you have already annotated your file with open-cravat). |
-o |
Path to your '.tsv' output file. |
-s |
You can provide a number of lines (variants), that you would like to see in your sorted output. |
-d |
Path to the folder with files required for annotation (default is 'db', located in the repo). |
-t |
Number of parallel processes. NOTE: Has high RAM requirements. |
-S |
If 'True', Create a separate file with predicted splice variants (SpliceAI > 0.7) in the same directory as output. |
| Module name |
|---|
| ClinVar |
| dbSNP |
| Fathmm |
| Fathmm_mkl |
| GenoCanyon |
| GERP |
| GnomAD v3.1.1 |
| InterPro |
| LRT |
| MetaLR |
| MetaSVM |
| MutationAssessor |
| MutationTaster |
| OMIM |
| Polyphen2 |
| Provean |
| SIFT |
| Siphy |
| SpliceAI |
| HPO |
These modules can be installed with the following command (make sure open-cravat is installed):
oc module install clinvar dbsnp fathmm fathmm_mkl genocanyon gerp gnomad3 interpro lrt metalr metasvm mutation_assessor
mutationtaster omim polyphen2 provean sift siphy spliceai hpo
Make sure, that you have all the base open-cravat modules installed.
oc module install-base
Feel free to report any issues, we are still testing our software. Any suggestions would be highly appreciated :)
Due to complicated installation of open-cravat and large size of its modules, we will try our best to create a docker image,
that is going to use open-cravat servers for annotation.
With the recent update of OpenCravat, the updated ClinVar annotator has been introduced, which interferes with the functionality of our software. To be more specific, the final output might have missing information for particular variants.
The easiest thing to do at the moment is to navigate to ClinVar annotator directory
(/OpenCravat_modules_folder/annotators/clinvar) and edit clinvar.py file by removing the
variant_clinical_sources from col_order list variable.
This section will be removed when we fix this issue.