Accurate prediction of collision cross sections (CCS) in proteomics is crucial for enhancing ion mobility spectrometry in mass spectrometry workflows. Here, the authors introduce a pretrained deep protein language model that significantly reduces training time while maintaining or improving CCS prediction accuracy, offering an efficient approach to proteomic analysis.
- Ayano Nakai-Kasai
- Kosuke Ogata
- Toshiyuki Tanaka
