Giulio Isacchini

excited to announce that this January we are relocating to Yale to start a theory initiative at the Center for Systems and Engineering Immunology (CSEI): @yalecsei.bsky.social

looking to hire 2 postdocs to join us in this endeavor.

more details can be found here: sites.google.com/uw.edu/statp...

📢 We are announcing the Adaptive Immune Profiling Challenge 2025!
Can you predict immune state labels from adaptive immune receptor repertoires?
💰 $10,000 prize pool!
🗓️ Launches Nov 5 on @kaggle.com
More Info: uio-bmi.github.io/adaptive_imm...

Thrifty wide-context models of B cell receptor somatic hypermutation Convolutional embedding models efficiently capture wide sequence context in antibody somatic hypermutation, avoiding exponential k-mer parameter scaling and eliminating the need for per-site modeling.

The final version of our "Thrifty" paper is up now: elifesciences.org/articles/10... .

We were motivated to fit wide-context mutation models based on previous analyses showing "mesoscale" effects and a position-specific effect. But, how to avoid exploding the number of parameters? 🧵

I'm stoked to have this long-term and highly collaborative effort out in the wild. We present an experimental evolution system to quantitatively study the rapid evolutionary dynamics of B-cell affinity maturation. These evolutionary mechanisms underpin antibody responses and immune memory formation.

Somatic hypermutation unlocks antibody specificities beyond the primary repertoire Initial specificity is thought to be required for somatic hypermutation (SHM)-driven antibody evolution. Zuo, Gautam, et al. demonstrate that non-cognate B cells can enter germinal centers, undergo SHM, and acquire new antigen specificities, highlighting the capacity of antibodies to generate new recognition through SHM.

Online now: Somatic hypermutation unlocks antibody specificities beyond the primary repertoire

Quantifying the evolutionary dynamics of structure and content in closely-related E. coli genomes Abstract. Bacterial genomes primarily diversify via gain, loss, and rearrangement of genetic material in their flexible accessory genome. Yet the dynamics

Thrilled to announce that our work (w. with fantastic @neher.io and Liam Shaw) has been published on Molecular Biology and Evolution! 🎉

academic.oup.com/mbe/advance-...

Are you curious about how fast the genome of E.coli evolves structurally (gains, rearrangements...) ? 🧬

A summary thread [1/N]🧵

After a long and winding odyssey, excited to finally drop anchor in open-access waters. This preprint shows how neutral allele frequency time series can illuminate disease transmission rates between communities— key for epidemic fore- & backcasting. medrxiv.org/content/10.1... 🧵

This work wouldn’t be possible without the help and support of my supervisors @thearmita.bsky.social, @thierrymora.bsky.social, and Aleksandra Walczak—thank you!

YouTube video by AIRR Community AIRR-C Seminar Series, Sep 26, 2024 - Giulio Isacchini, Imprint Labs, USA

Curious about generative modeling for T and B cell receptor repertoires and its role in understanding immune system diversity? Watch my talk here youtu.be/HVUYG-UmRH8?.... Thanks to the @airr-community.bsky.social and @victorgreiff.bsky.social for the opportunity to share our work!

Would like to join too!

meme paper summary showing a fork in the road for the tcr:pmhc specificity problem: 1. pointing to binary classification 2. pointing towards gen AI

My first Skeetorial!

💻🧬TCR-TRANSLATE - A new framework for thinking about the TCR:pMHC specificity problem.

TLDR:
We pretrained LLMs on ~8M TCR & pMHC seqs
Finetuned on sparse pMHC->TCR pair data
Validated CDR3b sequences to unseen antigens
>> random performance on IMMREP2023 "private" antigens

yes please!