I have obtained several hundred raw, unassembled yeast genomes from NCBI and I am looking for advice on how to process the genomes for downstream analysis.

I have a reference genome (S288C) to use for assembly/alignment but I'm not sure what to do first.

So far I have run a quality assessment (FastQC) followed by trimming (Trimmomatic). But I'm not sure if I should align or assemble next. The problem with yeast is if I assemble to a reference genome it reduces the genome to a haplotype and could miss moments of heterozygosity present. And I'm not sure how it would work to do a de novo assembly as I'm assuming it's the same problem. Same with aligning to the reference genome.

Are there ways to work around this? Does the yeast research community just ignore this issue? Are there tools currently out there that take all of this into account?


1 Answer 1


What is your data type? Many assemblers can resolve heterozygosity.

PacBio HiFi reads can be used with hifiasm to resolve both haplotypes rather trivially. This is the best case scenario and older sequencing projects are unlikely to have HiFi reads.

If they are merely PacBio subreads, you can use Falcon-Unzip to get heterozygous regions (though this is now quite old and I am not sure that it is recommended anymore).

For Illumina, the options are worse. Platanus is an assembler intended for the purpose of assembling heterozygous genomes. If nothing else, you can make the assembly and then just call variants.

Note that many yeast are probably sequenced from a haploid strain.

  • $\begingroup$ I have both PacBio HiFi reads as well as Illumina seqeunces. I'll defineityl try out hifiasm for the HiFi reads but I have a ton of downloaded Illumina sequences that I don't know how to process. If I were to perform a de novo assembly with SPAdes on those sequences and then called the variants with bcftools or something of the sort would that give me reliable data? $\endgroup$
    – rimo
    Sep 7, 2022 at 18:56
  • 1
    $\begingroup$ @rimo looking in more detail, there are better Illumina assemblers for highly heterozygous genomes. Updating answer with this info. $\endgroup$ Sep 7, 2022 at 19:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.