Yo Akiyama

This study shows that large protein language models can predict homo-oligomeric interface contacts, with performance improving with model size and alignment diversity.

This paper introduces MSA Pairformer, a memory-efficient protein language model that outperforms existing methods in contact prediction and variant effect prediction.

This study identifies molecular and immune features associated with response to anti-PD-(L)1 therapy in NSCLC using a cohort of 393 patients.