The OMG dataset: An Open MetaGenomic corpus for mixed-modality genomic language modeling From friends at Tatta Bio GitHub: github.com/TattaBio/OMGwww.biorxiv.org/content/10.1...
Biological language model performance depends heavily on pretraining data quality, diversity, and size. While metagenomic datasets feature enormous biological diversity, their utilization as pretraini...
Terrabacteria: redefining bacterial envelope diversity, biogenesis and evolution #NatureRevMicrowww.nature.com/articles/s41...
AntiDefenseFinder! And it is available also as an option with DefenseFinder: defensefinder.mdmlab.frwww.biorxiv.org/content/10.1...
bioRxiv - the preprint server for biology, operated by Cold Spring Harbor Laboratory, a research and educational institution
Methanogenesis outside the Euryarchaeota experimentally demonstrated by three cultivation-driven studies (two from my lab)! A long🧵.🐻with me tinyurl.com/4v4fkda6tinyurl.com/yr4p7js6tinyurl.com/mtsrj6b9
If you are interested in prophages, we have a new database: Prophage-DB. Check it out and all feedback is welcome biorxiv.org/cgi/content/...
bioRxiv - the preprint server for biology, operated by Cold Spring Harbor Laboratory, a research and educational institution
Phylogenetic reconciliation: making the most of genomes to understand microbial ecology and evolution #ISMEJournalacademic.oup.com/ismej/advanc...
A global atlas of soil viruses reveals unexplored biodiversity and potential biogeochemical impacts - Nature Microbiology www.nature.com/articles/s41...@apcamargo.bsky.social Are these new genomes already in the current version of IMG/VR or would I want to download separately and combine for now?
This study presents an extensive global compendium of metagenomically derived sequences that will serve as a foundation for understanding the role of viruses in soil ecosystems.
It's the most wonderful time of the year: the time when we all learn about the latest nanopore updates via screenshots of tweets of photos of slides Looks not disappointing, no sign of the accuracy wall! And dorado 0.7 is now out with the new models: github.com/nanoporetech...
I reviewed the AlphaFold3 paper from DeepMind for the journal Nature. I tried really hard to get the editors to demand that DeepMind release the code (even an executable) so people could do the many high-throughput studies we saw for AF2 (see image from my review). I failed. So just a server 4 now.