T2T

[robert-a-forsyth]

I'd like to run ClairSTO on samples mapped to Chm13 and I was wondering if I need to adapt ClairSTO to run successfully on them?

[aquaskyline]

Nothing special, you just need to use the correct reference genome as input to ClairS-TO.

[robert-a-forsyth]

Are there recommended PoN for chm13 to use?

[aquaskyline]

No, you will need to do liftover yourself. Basically, for somatic variant calling and cancer genome study, I suggest using GRCh38 as the reference, which is the mostly used reference genome for medically related workflows and databases.

[robert-a-forsyth]

I've been using lifted over resources from T2T consortium. I've been having some weird problems which I was hoping you might have some insight into

this is the command I ran

singularity exec \
  -B home/ \
  home/clairs-to_latest.sif \
  /opt/bin/run_clairs_to \
  --tumor_bam_fn tumor.bam \
  --ref_fn chm13v2.0.fa \
  --threads 24 \
  --platform hifi_revio_ssrs \
  --conda_prefix /opt/micromamba/envs/clairs-to \
  --panel_of_normals Homo_sapiens-GCA_009914755.4-2022_10-gnomad.vcf.gz,chm13v2.0_dbSNPv155.vcf.gz,1KGP.CHM13v2.0.whole_genome.recalibrated.snp_indel.pass.phased.native_maps.biallelic.3202.vcf.gz,CoLoRSdb.CHM13.v1.0.0.deepvariant.glnexus.vcf.gz?download=1" \
  --panel_of_normals_require_allele_matching 'True,True,False,False' \
  --output_dir ./out

and it ends up breaking at step 3 like so

[INFO] STEP 3: Non-somatic Tagging for SNV Variants
[INFO] RUN THE FOLLOWING COMMAND:
( parallel --joblog out/logs/parallel_3_nonsomatic_tagging_snv.log -j 8 pypy3 /opt/bin/clairs_to.py nonsomatic_tagging --pileup_vcf_fn out/tmp/vcf_output/snv_pileup.vcf --output_vcf_fn out/tmp/vcf_output/snv_nonsomatic_tagging_{1}.vcf --ctg_name {1}  --pypy3 pypy3 --parallel parallel --panel_of_normals home/Homo_sapiens-GCA_009914755.4-2022_10-gnomad.vcf.gz,home/chm13v2.0_dbSNPv155.vcf.gz,home/1KGP.CHM13v2.0.whole_genome.recalibrated.snp_indel.pass.phased.native_maps.biallelic.3202.vcf.gz,home/CoLoRSdb.CHM13.v1.0.0.deepvariant.glnexus.vcf.gz?download=1  --panel_of_normals_require_allele_matching True,True,False,False  :::: out/tmp/CONTIGS ) 2>&1 | tee out/logs/3_NONSOMATIC_TAG_SNV.log


gzip: stdout: Broken pipe
Killed

gzip: stdout: Broken pipe
Killed

gzip: stdout: Broken pipe
Killed
Killed

gzip: Killed
stdout: Broken pipe

I've attached the complete log file. Are these errors to do with formatting in the PON files? or the size of them perhaps? Any insight that could point me in the right direction towards solving the problem would be greatly appreciated. Thank you

clairsto_log.txt