#  SpecificityStudio preprint on bioRxiv! 

 





June 15, 2026

 

 

     ![Cartoon representing how results from a mutational library of DraNramp are predicted by context-free and context-aware sequence models](/sites/g/files/omnuum7206/files/styles/hwp_16_9__480x270/public/2026-06/Fig%20Gaudet%20Lab%20Website.png?h=a172e91c&itok=mjPhoWFc) 

 



 

A key result is that sequence models that learn from the surrounding sequence context - especially protein language models - systematically bias against variants that alter specificity. Read more [on bioRxiv](https://www.biorxiv.org/content/10.64898/2026.06.10.731299).



 

 

 



 

 

 Share on:- [     Facebook ](#)
- [     Twitter ](#)
- [     Linkedin ](#)