Hi Yunye
During liftover from hg19 to hg38, I noticed that a subset of variants cannot be successfully mapped using the standard hg19ToHg38.over.chain.gz file. Recently, UCSC released an updated auxiliary chain file, hg19ToHg38.chainBridge.over.chain.gz (Released 2026-03-13; https://hgdownload.cse.ucsc.edu/goldenPath/hg19/liftOver/ ), which can serve as a bridge to improve mapping coverage.
I would like to suggest a two-step liftover approach in GWASlab:
First step: use the standard hg19ToHg38.over.chain.gz to convert variants.
Second step: for variants that fail in the first step, attempt conversion using hg19ToHg38.chainBridge.over.chain.gz.
This workflow can significantly increase liftover success rates while maintaining the reliability of the standard chain. Supporting this as an optional feature in GWASlab would be very useful for users working with large-scale GWAS datasets.
Thank you very much for considering this suggestion.
Best regards,
Mengmeng
Hi Yunye
During liftover from hg19 to hg38, I noticed that a subset of variants cannot be successfully mapped using the standard hg19ToHg38.over.chain.gz file. Recently, UCSC released an updated auxiliary chain file, hg19ToHg38.chainBridge.over.chain.gz (Released 2026-03-13; https://hgdownload.cse.ucsc.edu/goldenPath/hg19/liftOver/ ), which can serve as a bridge to improve mapping coverage.
I would like to suggest a two-step liftover approach in GWASlab:
First step: use the standard hg19ToHg38.over.chain.gz to convert variants.
Second step: for variants that fail in the first step, attempt conversion using hg19ToHg38.chainBridge.over.chain.gz.
This workflow can significantly increase liftover success rates while maintaining the reliability of the standard chain. Supporting this as an optional feature in GWASlab would be very useful for users working with large-scale GWAS datasets.
Thank you very much for considering this suggestion.
Best regards,
Mengmeng