Ai-driven spectral decomposition: predicting the most probable protein compositions from surface enhanced raman spectroscopy spectra of amino acids

HIGHLIGHTS

• What: By training the network on the SERS profiles of 20 amino acids of human proteins the authors explore the feasibility of predicting the predominant proteins within the µL-scale detection volume of SERS. The authors explore whether this architecture, which has revolutionized natural language processing, holds the key to decoding the rich tapestries of SERS spectra. Originally designed to manage sequential data in natural language processing tasks, transformers brought the concept of "attention"-allowing the model to focus on different parts of the input data differently, akin to how humans pay attention to specific words or . . .
  

   

  

  If you want to have access to all the content you need to log in!

  Thanks :)

  Username or E-mail:

  If you don't have an account, you can create one here.