chhylp123 / hifiasm

Hifiasm: a haplotype-resolved assembler for accurate Hifi reads
MIT License
547 stars 87 forks source link

Question regarding quality threshold for PacBio and ONT assembly #689

Open s-t-calus opened 3 months ago

s-t-calus commented 3 months ago

Noticed that Illumina recommends: "For heterozygous calls, both alleles should have an allele-call score ≥ 10, and the ratio of their scores should be ≤ 3" - link

What allele threshold and ratio are used during the hifiasm assembly when processing PacBio HiFi and ONT R10.4.1 data?

Missing the allele-call score ≥Q10 generated us a lot of false positive SNP's in our Illumina data (DRAGEN pipeline), shall we be concerned about ONT errors inflating the number of variations in the sample? I can imagine assembly with PacBio data having a much higher allele-call score threshold and generating fewer false positives but what are the cut-off points?

All the best, S-T