I don’t have time to write much about it at the moment, but I wanted to note that there is a new Nature paper out today regarding the sequencing, assembly, and analysis of a few hundred new bird genomes:
The paper is short (possibly too short - I’d like to see a lot more detail without digging into the supplemental), open-access, and it touches on a bunch of topics of perennial interest.
So this is the direction things are going now? You have to publish 100s of new genomes in the same paper to be suitable for publicaton?
That reflects a pretty amazing advance the technology. Wow. Just a couple decades ago, a single new genome was big news.
It’s kind of strange going back and reading old papers on something like the amino acid sequence of a single protein.
In Nature? Pretty much, or something equivalently sexy/difficult/useful. In most journals? Certainly not.
Afraid so. Still, it’s getting to be that the cheapest way to sample lots of loci from lots of species is just to sequence the whole genome and dig the loci out of the data.
Way too short from my perspective. The methods section doesn’t even say anything about how the tree was generated. And the tree shown is illegible. This is at best the abstract of a real paper that analyzes the data for all sorts of questions. But that’s Nature for you.
True. Sequencing is cheap and easy. High quality assembly and annotation can be pain in the ass (though it’s a fun challenge). At least in plants - birds may have more compliant genomes.
Yeah, that makes sense @davecarlson. I’m just a bit awestruck about what is even possible now. Projects like this would have been inconceivably ambitious not long ago.
I think they do. Allopolyploidy is not so common in birds (may not exist) and gene duplication, which may be the same thing or may not, seems not as common either. Paralogy could be a big problem in assembly. Also bird genomes seem to maintain synteny for a very long time, they’re small genomes as vertebrates go, and they have fewer big eruptions of transposable elements than many other vertebrates.
Give it another year or two and some middle school kid’s science project will be titled, “101 Genome Sequences from the Flora and Fauna I Found in My Dog’s Nose.”
That paper is one of a triplet published today, all on the subject of multi-genome alignments:
I think the one describing the tool that allows for these massive comparisons is the most interesting:
https://www.nature.com/articles/s41586-020-2871-y
Oh look, a similar paper with ~120 new mammal genomes also published in Nature today:
2 posts were split to a new topic: Genomic Research in Africa by Africans
It turns out (talking to one of the authors) that the tree in the paper and the supplementary material doesn’t even come from the data. It’s just a supertree assembled from previous publications, and preliminary analyses show no real support for many of the branches. I’m bummed.
Agreed. There has been a Moore’s Law-like increase in DNA sequencing and bioinformatics since the 1980’s. Within the modern field of genetics, programming skills are as important as bench skills. More than once in the last year I have said something to the effect of “Why look at the expression of just a handful of genes when we can look at all of them at once at once?”. It’s nuts.
Bioinformatics is a field young biologists should be seriously considering, as these sets of papers demonstrate.
It’s certainly what I spend most of my time doing (mostly by my choice).
When I started out in grad school in 1990, I was radiolabeling ddNTPs, doing 4 sequencing reactions by hand, one for each base, and exposing the dried gels overnight on x-ray film. Now that was tedious. In my last year, the lab got an auto-sequencer and we started using dye-labeled ddNTPs and cycle sequencing. The gel sequencer was eventually replaced by a tube sequencer, and much later by an Illumina machine that cranks out whole genomes. Would have been nice.