Does anybody know what’s the advantage or disadvantage in using reads to infer phylogenies instead of using assemblies?
I´ve been searching for information about that question and I only found this “For SNP discovery, using raw reads can provide greater resolution than using a genome assembly. With raw reads, both the depth of coverage as well as the proportion of mixed alleles can be quantified, in contrast to creating an assembly, in which all coverage at a given locus is collapsed into a single base call. When the raw reads are available they can be mapped back to the assembly to obtain the coverage and allelic proportions. However, lack of a closed reference genome may cause biases in allelic proportions due to mapping errors”
But that was all I got. I tried to find more information using the references in that article but it gave me the same information. If you have some information about that, I will be so grateful!
I’ve never heard of anyone doing real phylogenetics from reads. That doesn’t make a lot of sense to me.
Do you have a link or a paper or something you’re basing this off?
This is the link
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4493402/
They used reads to identify and evaluate SNP calling
Thanks
Sounds like they didn’t use the reads per se, but instead called SNPs relative to a reference or de novo assembly.
You may want to follow something like the steps described here: https://www.ncbi.nlm.nih.gov/pubmed/22886560 . First call SNPs from the reads, then build a SNP tree. This is a little different to just using the reads directly.
It has been some time since I've worked with phylogenies, and I've used the "traditional" method of assembling, aligning and reconstructing phylogeny with ML / Bayesian approach, so I don't have hands-on experience with the methods you are asking about. However, this paper seems like an answer to your question: Alignment-free sequence comparison: benefits, applications, and tools