RT @CSHL: Developed by CSHL researchers, BATMAN is a new AI model trained to predict which cancer markers immune cells will target—while av….

Understanding cross-reactivity is essential for TCR based therapies and we hope BATMAN as a stand alone tool, plus BATCAVE for future methods development can get us one step closer to this goal.

Join millions who have switched to Grok.

We demonstrate the translational potential of BATMAN by accurately ranking immunogenic peptides of clinically relevant TCRs, essential for understanding neoantigen immunogenicity and off-target effects of TCR-based therapies.

We then developed an active learning framework for BATMAN, to minimize the number of TCR-pMHC exps to be performed to learn the xreactivity of a novel TCR, thus guiding experimental biologists in the optimal strategy for designing experiments to determine TCR specificity.

BATCAVE became our benchmark to develop BATMAN, a probabilistic framework to predict TCR-pMHC activation and cross-reactivity. BATMAN learns interpretable features such as us AA-distance functions and epitope positional weights to correlate with biophysical TCR-pMHC properties.

For BATCAVE, we scoured the TCR-pMHC lit from the past 20 years, collecting any mutational scan data we could get our hands; we processed these in a common framework, obtaining a databases of > 22k+ TCR-pMHC activation datasets across human and mouse, CD4 and CD8 epitopes.

The work was a great collaborative effort between my lab and @navlakha_lab at @CSHL largely driven by.Amitava Banerjee, David Pattinson. Curious about BATCAVE and BATMAN? Read on ->.

Check out how we tackled one of the fundamental challenges in immunology, out today.@CellSystemsCP: Our major innovations are BATCAVE and BATMAN, two new players to get us closer to solving TCR-pMHC predictions!.

Our latest work with the @JanowitzTobias lab @CSHL is out today - we used latent trait genomics to uncover cystatin C's role in signaling and cancer immunotherapy! Fantastic collaborative effort summarised below by Sam.

The thymus is going public @Radiolab! Great story on the biology behind your inner mini-self and the ignore list it equips your immune system with: Featuring our work and live scenes with @CarcySalome from our lab @cshl!.

T cell education is our favourite puzzle in biology - come join us @cshl and help us solve the mysteries of promiscuous gene expression and the mechanisms involved! We have 2x5y funded postdoc positions, immunology/comp bio background welcome! Contact me for more details! .10/10.

Thank you to everyone involved: Jason Carter, Léonie Strömich, Sarah Chapin, @matty_peacey @LarsMSteinmetz, @larsplus, @bb_610 and Sheena Pinto. This work was inspired by many discussions with our mentor and colleague, the late Dr Bruno Kyewski. 9/10.

To browse the 'mini you' that T cells get to see, check out our interactive public interface for exploring mTEC transcriptomic diversity at 8/10.

Our results help construct a map of transcriptomic diversity in the healthy human thymus and may ultimately facilitate the identification of transcripts and epitopes which contribute to autoimmunity and immune recognition of tumor antigens. 7/10.

upregulated during mTEC maturation, but contribution to initiation of transcription is rare, suggesting that ERE expression is largely a passive consequence of promiscuous gene expression. 6/10.

Key players in promiscuous gene expression are AIRE and FEZF2. We found increased rates of global transcript mis-initiation with transcript initiation stochasticity prevalent among AIRE, but not FEZF2 genes. In addition, long terminal repeat retrotransposons are . 5/10.

protein coding genes detected across our samples. Mechanistically, this seems to be driven primarily by differential expression of peripheral splicing factors rather than generating novel transcripts through promiscuous splicing. 4/10.

their training and thereby learn what’s self and harmless. We studied how complete this ignore list is in humans and how these epithelial cells manage to generate it. We found that on average human mTECs promiscuously express more than 78k transcripts, with 96% of all . 3/10.

T cells need extensive training before being sent on their quest to fight infections. To ensure that these fights don’t result in auto-immune reactions, medullary thymic epithelial cells (mTECs) generate an ignore list of peptides found in your body that T cells see during . 2/10.

How are T cells trained to distinguish healthy self from harmful foreign? In our recent paper just out in @NatureComms we studied the ‘mini you’ that T cell educators in the thymus generate to teach T cells about the healthy self. 1/10.