Support for telomere to telomere build - T2T - CHM13v2

[davmlaw]

Hamish asked whether we support this. Ensembl have released VEP that supports it now

Ensembl/ensembl-vep#1409

Working in feature/t2t_genome_build

[davmlaw]

VEP fasta:

wget --quiet -O - https://ftp.ensembl.org/pub/rapid-release/species/Homo_sapiens/GCA_009914755.4/ensembl/genome/Homo_sapiens-GCA_009914755.4-softmasked.fa.gz | gzip -d | bgzip > Homo_sapiens-GCA_009914755.4-softmasked.fa.gz
samtools faidx Homo_sapiens-GCA_009914755.4-softmasked.fa.gz

VEP

wget https://ftp.ensembl.org/pub/rapid-release/species/Homo_sapiens/GCA_009914755.4/ensembl/variation/2022_10/indexed_vep_cache/Homo_sapiens-GCA_009914755.4-2022_10.tar.gz
# This needs to be renamed/copied to "homo_sapiens"
mkdir -p homo_sapiens  # in case already exists w/Ensembl
mv homo_sapiens_gca009914755v4/107_T2T-CHM13v2.0 homo_sapiens
rmdir homo_sapiens_gca009914755v4

Fasta

wget --quiet -O - https://ftp.ncbi.nlm.nih.gov/genomes/all/GCF/009/914/755/GCF_009914755.1_T2T-CHM13v2.0/GCF_009914755.1_T2T-CHM13v2.0_genomic.fna.gz | gzip -d | bgzip > GCF_009914755.1_T2T-CHM13v2.0_genomic.fna.gz
samtools faidx GCF_009914755.1_T2T-CHM13v2.0_genomic.fna.gz

Liftover

wget https://hgdownload.soe.ucsc.edu/goldenPath/hs1/liftOver/hs1ToHg19.over.chain.gz https://hgdownload.soe.ucsc.edu/goldenPath/hs1/liftOver/hs1ToHg38.over.chain.gz https://hgdownload.soe.ucsc.edu/goldenPath/hg19/liftOver/hg19ToHs1.over.chain.gz https://hgdownload.soe.ucsc.edu/goldenPath/hg38/liftOver/hg38ToHs1.over.chain.gz

VEP annotation files

https://hgdownload.soe.ucsc.edu/gbdb/hs1/ - lots of things here - see UCSC browser T2T to see track listings

Clinvar

wget https://ftp.ensembl.org/pub/rapid-release/species/Homo_sapiens/GCA_009914755.4/ensembl/variation/2022_10/vcf/2024_07/clinvar_20240624_GCA_009914755.4.vcf.gz
python manage.py clinvar_import /data/annotation/variantgrid_setup_data/clinvar/t2t/clinvar_20240624_GCA_009914755.4.vcf.gz

gnomAD

Downloading v4 exomes + genomes from here:

https://ftp.ensembl.org/pub/rapid-release/species/Homo_sapiens/GCA_009914755.4/ensembl/variation/2022_10/vcf/2024_07/

Will need to combine them

Somalier

wget https://github.com/brentp/somalier/files/9954286/sites.chm13v2.T2T.vcf.gz

cdot

wget https://github.com/SACGF/cdot/releases/download/data_v0.2.27/cdot-0.2.27.ensembl.T2T-CHM13v2.0.json.gz https://github.com/SACGF/cdot/releases/download/data_v0.2.27/cdot-0.2.27-Homo_sapiens_T2T-CHM13v2.0_Ensembl_2022_06.gtf.json.gz

# all transcripts (not sure if has any novel ones over release below)
python3 manage.py import_gene_annotation --genome-build=T2T-CHM13v2.0 --annotation-consortium=Ensembl --json-file cdot-0.2.27.ensembl.T2T-CHM13v2.0.json.gz 
# Then the release
python3 manage.py import_gene_annotation --genome-build=T2T-CHM13v2.0 --annotation-consortium=Ensembl --release T2Tv2_Ensembl_2022_06 --json-file /data/annotation/cdot/Ensembl/T2T-CHM13v2.0/cdot-0.2.27-Homo_sapiens_T2T-CHM13v2.0_Ensembl_2022_06.gtf.json.gz

CONSERVATION

No phastcons/phylop but there is Cactus:

https://hgdownload.soe.ucsc.edu/gbdb/hs1/hgCactus/t2tChm13.v2.0.hal

Cactus reference-free alignments of GRCh38 and T2T CHM13 v2.0, using chimp (GCF_002880755.1/panTro6) as an out-group.

Data comes in HAL: a hierarchical format for storing and analyzing multiple genome alignments - I think you can convert it to phyloP type scores or to bigWig (hal -> halPhyloP -> wig -> bigWig)

REPEATS

repeat masker - https://hgdownload.soe.ucsc.edu/gbdb/hs1/t2tRepeatMasker/

dbNSFP - N/A checked site
dbSCSNV - N/A - but 38 is just lifted over from 37 anyway
COSMIC - N/A checked site
MAVE - N/A
Mastermind - ?
splice AI SNV/indel - ?
topmed - ?
uk10k - ?

Write up how to add genome build

Assembly/contig

wget https://ftp.ncbi.nlm.nih.gov/genomes/all/GCF/009/914/755/GCF_009914755.1_T2T-CHM13v2.0/GCF_009914755.1_T2T-CHM13v2.0_assembly_report.txt # Copied into source

A test VCF sample:

https://ftp-trace.ncbi.nlm.nih.gov/ReferenceSamples/giab/data/AshkenazimTrio/analysis/NIST_HG002_DraftBenchmark_defrabbV0.018-20240716/CHM13v2.0_HG2-T2TQ100-V1.1.vcf.gz

[davmlaw]

TODO

• gnomAD v4 is downloading on screen in vg.com once downloaded, combine then test out, set as v4 in T2T annotation VEP config
• Test annotation in old builds etc still works
• Find other data?
• Repeat masker bigbed?
• Still need to remove columns not in T2T - I think this is OK for VEP imports, but not for the annotation details page

In [17]: for cvf in qs.filter(q).order_by("source_field").filter(Q(vep_custom__isnull=False) | Q(vep_plugin__isnull=False)):
    ...:     print(cvf, cvf.vep_info_field)

[davmlaw]

View classification page, throws:

• No cached column for genome build T2T-CHM13v2.0

https://app.rollbar.com/a/jimmy.andrews/fix/item/VariantGrid/5585??utm_source=rollbar-notification