Back to All

GRAF benchmarking

Hey!
Hope you're well! We ran the GRAF pipeline for NA12878 with this inputs and take the ouput and ran hap.py comparing the results with the gold standard downloaded from this link: https://ftp-trace.ncbi.nlm.nih.gov/giab/ftp/release/NA12878_HG001/latest/GRCh38/ and obtain: 56.42 % of precision for all indels and 84.01% of precision for all snps. Based on the publication of GRAF and what we've seen on the blog, the expected precision was near 99% but we are not being able to reproduce these results. Do you know what could be wrong? Thanks in advance for your help!
Best,
Maria Teresita Laguinge