"340 years and they still haven't figured out their audio setup." LOL www.youtube.com/watch?v=db36...

Sequenced genomes from a recent EVD outbreak "form a well-supported phylogenetic cluster with genomes from the 1976 Yambuku/Mayinga outbreak"! www.researchsquare.com/article/rs-7...

Lucky, I was not drinking my morning coffee while reading this.

As expected, unfortunately. If ever you needed a reason for never using GISAID ever again (as a data producer or data user - we're both), look no further. Time to move on to more trusted and transparent solutions.

"...spike protein of this novel bat coronavirus possesses a functional furin cleavage site at the S1/S2 junction with a unique amino acid sequence motif (RDAR) that differs from that found in SARS-CoV-2 (RRAR) by only one amino acid..." www.biorxiv.org/content/10.1...

Interesting paper: "...no clear association between genomic or phenotypic traits and recombination rate...", but "the devil is in the details" doi.org/10.1073/pnas...

When Apple iCloud relay is turned on, you use Safari to browse Google scholar, Google blocks you. But when you switch to Google Chrome, no block... hmm, evil? hmm...

Turns out NLP is just vision

It is scandalously expensive now. Not to mention if you include your family. Let's hope British science can back it up in the long run.

So this is not a joke. A changing environment makes all the difference.

Such an "old" paper, but still gives insights: onlinelibrary.wiley.com/doi/10.1111/...

Ah cool! If we keep looking many seemly complex "molecular machineries" have originated and were even "perfected" before eukayotes.

Such a overrated LLM for debugging research code... www.swebench.com

One evolved with modern humans, the other one has been evolving from ~4 billion years ago. We need a different approach...

BBSRC responsive mode was such a nice scheme, would like to try again one day.

Should be helpful for building plasmid genetic models for AMR evolution: www.science.org/doi/10.1126/...

A nice paragraph! "Viral emergence and pandemic preparedness in a One Health framework" www.nature.com/articles/s41...

Using more evolutionary information is the right strategy.

This is crazy, I think I may have accidentally activated this model API call and then cancelled it. All input tokens, what a waste!

Without “unfunded hobbyist research” or in a sense "curiosity-driven research", the science as we know it will be lost. There is nothing left to inspire future generations of scientists.

How do loci for bacterial capsules (Klebsiella) manage to get swapped around despite their outstanding phenotypes? They just plug and play, having little impact on the rest of the genome. Kudos to @julielebris.bsky.social , check her thread below

Fascinating! virological.org/t/the-16th-e...

Very interesting, testing mutational hazard hypothesis in island red devil spiders. academic.oup.com/mbe/advance-...

Could smaller "antibodies" that work together like legos help us control evolution? www.nature.com/articles/s41...

Saying an LLM lies is anthropomorphizing it, and yet it's the most appropriate word for an experience that is quite common with these types of models.

Nice use of "cost of complexity". link.springer.com/article/10.1...

Not sure how many scientists here have tried Claude Code or similar command line coding assistants. I had a complicated family property tax problem that was best solved by a brute force Monte Carlo simulation approach, so I spent a few days coding up and analyzing a model with Claude Code.

Attempts to predict virus evolution using AI, but as authors and other pointed out they can be improved by combining some mechanistic modelling, e.g., epistasis... www.pnas.org/doi/10.1073/... www.nature.com/articles/s41... clauswilke.substack.com/p/limitation... www.pnas.org/doi/10.1073/...

This is interesting, but then, for those well aligned protein sequences (good evolutionary conservation), even the old methods, such as pairwise correlation-based methods would behave well. So evolutionary context would help improve future methods (scaling issues). www.biorxiv.org/content/10.1...

This is prob going to work to some good extent as the viral env in this context is produced by host cells, which is known to be extensively glycosylated in its native state.

Pacman!

Such a nice paper, there is a comparison to BEAST method. Note this group of methods is generally applicable to other types of clonal evolution. But recombination can make things complicated. doi.org/10.1093/gene...

This is pretty cool! FigTree can still handle a genealogy of two million cells.

This is interesting, particularly Lokiarchaea placement in fig 4, and application to TB clustering. More detailed comparisons with current and classic phylogenetic methods and application contexts would also be nice as authors already realised... www.biorxiv.org/content/10.1...

What's missing? www.cell.com/cell/fulltex... www.youtube.com/watch?v=_2Ni...

Predicting the Evolutionary and Functional Landscapes of Viruses with a Unified Nucleotide-Protein Language Model: LucaVirus www.biorxiv.org/content/10.1...

"Early Microbial Life: Our Past, Present, and Future" could/should make it onto your reading list for this summer. it's on ours. #MicroSky #Archaea #ArchaeaSky #ProtistsOnSky typo alert in FIG.17: "possed" > possessed

Sponges might also be interesting. www.cell.com/current-biol...

Wiredu-Boakye et al. analysed 432 DNA damage response (DDR) proteins in 68 genomes, showing that DDR proteins are consistently longer than housekeeping proteins, regardless of life history. 🔗 doi.org/10.1093/gbe/evaf089 #genome #evolution #DNA

H9N2 circulating in domestic poultry has the ability to transmit in experimental animals. www.nature.com/articles/s41...

If you don't pasteurise it, how long can H5N1 influenza viruses stay infectious for in milk? We had a look at this in a new preprint with ‪@cvrinfo.bsky.social‬, @roslininstitute.bsky.social‬, ‪@pirbrightinst.bsky.social‬ & the APHA (1/N) www.medrxiv.org/content/10.1...

Just got hit with this wave, not pleasant. It seems the spatially changing immune selection indeed modulates recombinants landscape.

Remember that Glasgow paper about discovering a new evolutionary law? There are similar symptoms in recent AI papers in explaining learned evolutionary patterns/mechanisms. www.pnas.org/doi/10.1073/...

Congrats team! The message is important: despite extensive recombination history, the pandemic strains need changing environmental factors to move them fast and wide before human emergence and pandemic, although it is challenging and only rarely successful... doi.org/10.1016/j.ce...

This is interesting, would this affect some famous databases such as ESM Metagenomic Atlas, and some downstream analyses, such as novel domain combinations?

"Immunization is a ‘best buy’ in health with a return on investment of $54 for every dollar invested and provides a foundation for future prosperity and health security." www.who.int/news/item/24...

A very interesting recent study. It seems the strategy that using synthetic data to develop and train deep learning models could work, particularly "where the generating process is understood but inference is challenging" (e.g., CNAs calling in this case). www.biorxiv.org/content/10.1...

So happy, I guess this is one of those "aha" moments in your research.

Interesting, applied for dating a notorious region, HLA, under selection, wonder how this improvement could be applied more broadly. www.biorxiv.org/content/10.1...