There's a typo on line 225, it is actually 50% identity, not 90%. But yeah, SRA is highly redundant :)

@martinsteinegger.bsky.social‬, @caleblareau.bsky.social, @pierrepeterlongo.bsky.social, @rnalab.bsky.social

@tlemane.bsky.social‬, @mmontonerin.bsky.social, @apcamargo.bsky.social, @mattlabguy.bsky.social, @sinamajidian.bsky.social‬, @rfaure.bsky.social‬, @jmouradesousa.bsky.social‬, @epcrocha.bsky.social‬, @david-koslicki.bsky.social, @pashadag.bsky.social‬,

Earth’s genetic diversity is a heritage of humanity. It has been an honour to explore this data with a team of dedicated scientists who shared our vision of making this data free and accessible to all 🌍🧬❤️ Thank you!

Updated preprint: doi.org/10.1101/2024...

This is a new frontier for biological discovery and AI training data. Logan expands the universe of known proteins, plasmids, AMR, P4 satellites, and the newly discovered Obelisk RNA elements.

All Logan data is freely-available (cc0) right now. We show how Logan-Search (www.logan-search.org) can be used to uncover viral reactivation (HHV-6) in cell therapy products (TIL and CAR-T).

Logan rapidly accesses the tapestry of Life’s genetic diversity and can help solve global issues.

To tackle the microplastic crisis, we searched Logan for new versions of the 213 known plastic-degrading enzymes. We identified 200+ million homologs 🤯, including new high-efficiency enzymes 🥤🔥

Logan enables minute-scale k-mer search, and hour-scale deep homology protein alignment search, across 100+ Billion proteins.

www.logan-search.org

One year after our initial preprint, we're excited to post a major update to Logan.

At its heart, Logan is the assembly of 27 million samples (50 Pbp) using a 6-day cloud-compute peaking at 2.2M vCPUs. This compresses the SRA 140x compared to raw FASTQs.

github.com/IndexThePlan...

thanks Niema!

thanks Ben!!

merci Sophie!

yes😅

For more context: Logan is a collection of all public sequencing data (until end of 2023) assembled into contigs. It is freely hosted on the cloud, and contains hundreds of terabytes of valuable genomic data: github.com/IndexThePlan...

Ty Rob!