Problems with vep annotation

[vedellpt]

I am trying to run tapes using a vep vcf as input. The vep vcf that I am using as input contains the ClinVar significance information in the CLIN_SIG entry which is part of the CSQ INFO field which has description "Consequence annotations from Ensembl VEP".

I have process vep vcfs for 37 patients. I know some have some pathogenic variants but none are classified as such by tapes. I think it probably has to do with this message in the log:

"2020-03-20 09:51:55.....PS1 done || No trio data, skipping PS2 || No "clinvar_golden_stars" or "clinvar_clnsig" column found. Please annotate your data with a recent clinvar database || All frequency data not found || No domain data found 2020-03-20 09:51:55.....PM5 done"

I think it could also be related to this message:

"2020-03-20 09:51:55.....Starting... || Cannnot calculate PVS1, no splicing annotation. Please annotate with dbscSNV"

Can you please provide some suggestions on how I can get past these problems? Thanks.

[vedellpt]

I have been able to get db, annotate, and sort to work on the toy dataset that you have as an example. I have tried a number of things and have continued to be unsuccessful in getting it a successful run of annotate on my annovar input vcf and I also continued to be unsuccessful in getting a successful run of sort on my vep input vcf. Is there anything different about the way the toy example dataset processing is done compared to that of other datasets?

[vedellpt]

For my latest effort, I take a vep vcf from the vep web portal and try to use it as input to tapes sort. I get the same error for an ascii, Unicode and Unix formatted input file. It is as shown below. Can you tell me how to get past this? Is there a way to tell it which format it is? It seems that it is unable to figure it out in this function call.

[m4@mf4 tapes 2020-03-23 02:18:27] $ cat /research/tapes_sort_veptest8.err Traceback (most recent call last): File "tapes.py", line 355, in main() File "tapes.py", line 205, in main full_stuff, soft_used = tf.open_csv_file(file_path, acmg_db_path) # Load annotated csv in pandas File "/research/TAPES/tapes/src/t_func.py", line 63, in open_csv_file soft_used = check_which_soft_used(csv_file) File "/research/TAPES/tapes/src/t_func.py", line 3723, in check_which_soft_used for line in input_vcf.readlines(): File "/research/python/3.6.7/lib/python3.6/encodings/ascii.py", line 26, in decode return codecs.ascii_decode(input, self.errors)[0] UnicodeDecodeError: 'ascii' codec can't decode byte 0xff in position 0: ordinal not in range(128)

[vedellpt]

Anyway, congratulation on your Plos Computational Biology publication and on developing this tool. I think it is a nice publication and I think it serves an important purpose.

[EugeneEA]

Anyway, congratulation on your Plos Computational Biology publication and on developing this tool. I think it is a nice publication and I think it serves an important purpose.

I also have some problems with VEP vcf annotation, but as far as it seems that tool does not have a support I'm not sure that it makes sense to open an issue ( (I'd like to be wrong, because there are few options for ACMG assignments :( )

[NTNguyen13]

Hi, I found the way to solve this problem:

You will need 2 plugins: dbscSNV and dbNSFP for all annotations, add it to VEP command like this:

    --plugin dbNSFP,/path/to/dbNSFP_hg19.gz,gnomAD_genomes_AF,gnomAD_exomes_AF,CADD_phred,FATHMM_converted_rankscore,clinvar_clnsig,clinvar_golden_stars,Interpro_domain,SIFT_score,LRT_pred,MutationTaster_pred,MutationAssessor_pred,FATHMM_pred,PROVEAN_score,MetaSVM_pred,MetaLR_pred,M-CAP_pred,fathmm-MKL_coding_pred,GenoCanyon_score,GERP++_RS \
    --plugin dbscSNV,/path/to/dbscSNV1.1_GRCh37.txt.gz

[EugeneEA]

Hi, I found the way to solve this problem:

You will need 2 plugins: dbscSNV and dbNSFP for all annotations, add it to VEP command like this:

    --plugin dbNSFP,/path/to/dbNSFP_hg19.gz,gnomAD_genomes_AF,gnomAD_exomes_AF,CADD_phred,FATHMM_converted_rankscore,clinvar_clnsig,clinvar_golden_stars,Interpro_domain,SIFT_score,LRT_pred,MutationTaster_pred,MutationAssessor_pred,FATHMM_pred,PROVEAN_score,MetaSVM_pred,MetaLR_pred,M-CAP_pred,fathmm-MKL_coding_pred,GenoCanyon_score,GERP++_RS \
    --plugin dbscSNV,/path/to/dbscSNV1.1_GRCh37.txt.gz

Hi, thanks for posting it. Eventually I've exchanged VEP for OpenCravat and re-write the code of InterVar to make both compatible.

[land-mine]

Hello @EugeneEA, I'm working on the variant interpretation. If you can provide the modified code of InterVar it will be helpful.

[SouzaBB]

Hi, I found the way to solve this problem:

You will need 2 plugins: dbscSNV and dbNSFP for all annotations, add it to VEP command like this:

    --plugin dbNSFP,/path/to/dbNSFP_hg19.gz,gnomAD_genomes_AF,gnomAD_exomes_AF,CADD_phred,FATHMM_converted_rankscore,clinvar_clnsig,clinvar_golden_stars,Interpro_domain,SIFT_score,LRT_pred,MutationTaster_pred,MutationAssessor_pred,FATHMM_pred,PROVEAN_score,MetaSVM_pred,MetaLR_pred,M-CAP_pred,fathmm-MKL_coding_pred,GenoCanyon_score,GERP++_RS \
    --plugin dbscSNV,/path/to/dbscSNV1.1_GRCh37.txt.gz

Hi... I'm running tapes using my VEP output just like you've posted here but, still having isuue with error: All required annotations not found.

Did you managed to solve the issue??