Fantastic project led by @bo-wen.bsky.social. Excited to see the future uses of AI and transfer learning in proteomics. #massspec #proteomics www.nature.com/articles/s41...

DeepMVP is a deep learning framework to predict PTM sites and variant-induced alterations across 6 common PTMs. www.nature.com/articles/s41...

Cascadia from @wnoble.bsky.social is a mass spec-based de novo sequencing model that uses a transformer architecture to handle data-independent acquisition data and achieves substantially improved performance across a range of instruments and experimental protocols. www.nature.com/articles/s41...

(1/2) We have always validated FDR internally on several datasets. Bo Wen and colleagues discovered that FDR of DIA-NN 1.8.1 was anti-conservative on some (but not other) datasets - for some unknown reason. So we fixed it in 2.0 :) Now q-values are more accurate and fluctate less across datasets.

Error control in proteomics mass spectrometry analysis is hard. We came up with a way to evaluate error control. Upshot: for old-school DDA data, not so bad. For DIA data, no existing tool successfully controls the false discovery rate! www.nature.com/articles/s41...

Excited to see this published! It is a good step in the process for people to assess their FDR control in proteomics experiments. Great work from @bo-wen.bsky.social and @urikeich.bsky.social in particular who drove this.

Assessing error control is fundamental in mass spectrometry-based proteomics. @bo-wen.bsky.social @maccoss.bsky.social @urikeich.bsky.social et al introduce a theoretical foundation for entrapment along with a method for more accurate evaluation of FDR control. www.nature.com/articles/s41...