Peter DeWeirdt

A membrane-bound nuclease directly cleaves phage DNA during genome injection - Nature A widespread bacterial defence system called SNIPE is shown to localize to the cell membrane, where it identifies and cleaves the DNA of infecting phage as it is injected into the bacterial cell.

Nature research paper: A membrane-bound nuclease directly cleaves phage DNA during genome injection

go.nature.com/3P2jKFh

👋 Hi antiphage defense community, we are soon releasing an update of DefenseFinder.

We are doing our best to include all the great discoveries from the community, but with so much going on we might miss things.

Please answer wt preprints/papers with new systems or mail/git them to us. 🙏

Protein and genomic language models chart a vast landscape of antiphage defenses The bacterial pangenome encodes an immense array of antiphage systems, yet much of their diversity remains uncharted. In this study, we developed language models to predict novel antiphage proteins in...

Also congrats to @audeber.bsky.social and co. on their work using ML to predict anti-phage defense systems: www.biorxiv.org/content/10.1...

8/8

DefensePredictor: A Machine Learning Model to Discover Novel Prokaryotic Immune Systems Anti-phage defense systems protect bacteria from viruses. Studying defense systems has begun to reveal the evolutionary roots of eukaryotic innate immunity and produced important biotechnologies such ...

Congrats Mike Laub and Emily Mahoney!

Lots more in the manuscript: www.biorxiv.org/content/10.1...

7/n

When applied to 500 diverse bacterial assemblies, DefensePredictor revealed thousands of additional candidate systems awaiting validation.

6/n

Among the systems were 7 protein domains that we newly validated in defense, including in DS-8 and DS-9, which have homology to the human innate immune protein SMPDL3A.

5/n

We tested the model’s predictions in lab and identified 45 novel defense systems, observing a strong correlation between the validation rate and defense probability.

4/n

DefensePredictor demonstrates strong performance on a held-out set of defense genes and leverages biological patterns like low GC content to make its predictions.

3/n

We compiled a dataset of known defense systems and trained a gradient boosting model to distinguish defense from non-defense genes, using ESM2 embeddings and genomic features of each gene and its neighbors.

2/n

Excited to share our work using machine learning to predict anti-phage defense systems!

1/n