Genesis: A modular protein language modelling approach to immunogenicity prediction

Neoantigen immunogenicity prediction is a highly challenging problem in the development of personalised medicines. Low reactivity rates in called neoantigens result in a difficult prediction scenario with limited training datasets. Here we describe Genesis, a modular protein language modelling approach to immunogenicity prediction for CD8+ reactive epitopes. Genesis comprises of a pMHC encoding module trained on three pMHC prediction tasks, an optional TCR encoding module and a set of context specific immunogenicity prediction head modules. Compared with state-of-the-art models for each task, Genesis’ encoding module performs comparably or better on pMHC binding affinity, eluted ligand prediction and stability tasks. Genesis outperforms all compared models on pMHC immunogenicity prediction (Area under the receiver operating characteristic curve=0.619, average precision: 0.514), with a 7% increase in average precision compared to the next best model. Genesis shows further improved performance on immunogenicity prediction with the integration of TCR context information. Genesis performance is further analysed for interpretability, which locates areas of weakness found across existing immunogenicity models and highlight possible biases in public datasets. ### Competing Interest Statement S.A.Q. is co-founder and chief scientific officer and own shares in Achilles Therapeutics. C.S. acknowledges grants from AstraZeneca, Boehringer-Ingelheim, Bristol Myers Squibb, Pfizer, Roche-Ventana, Invitae (previously Archer Dx Inc., a collaboration in minimal residual disease sequencing technologies), Ono Pharmaceutical and Personalis. He is chief investigator for the AZ MeRmaiD 1 and 2 clinical trials and is the steering committee chair. He is also co-chief investigator of the NHS Galleri trial funded by GRAIL and a paid member of GRAIL's scientific advisory board (SAB). He receives consultant fees from Achilles Therapeutics (also an SAB member); Bicycle Therapeutics (also an SAB member); Genentech; Medicxi; the China Innovation Centre of Roche (CICoR), formerly Roche Innovation Centre Shanghai; Metabomed (until July 2022); Relay Therapeutics; and the Sarah Cannon Research Institute. C.S has received honoraria from Amgen, AstraZeneca, Bristol Myers Squibb, GlaxoSmithKline, Illumina, MSD, Novartis, Pfizer and Roche-Ventana; previously held stock options in Apogen Biotechnologies and GRAIL; currently has stock options in Epic Bioscience and Bicycle Therapeutics; and has stock options and is co-founder of Achilles Therapeutics. S.R.H. is the cofounder of PokeAcell and is co-inventor of licensed patents related to T cell detection. H.O'B., M.S., L.M. and F.O'F. are employees of Achilles Therapeutics