Peptide-Spectra Matching from Weak Supervision
This work addresses a fundamental bottleneck in proteomics for biologists by enabling more accurate protein identification from limited ground truth data, though it is incremental as it builds on existing methods with a novel training approach.
The paper tackled the problem of identifying proteins from mass spectrometry data by improving peptide-spectra matching using deep neural networks trained on weak supervision from imperfect hand-coded models, achieving a 43% improvement over standard methods and up to a 15% advantage in more challenging experimental regimes.
As in many other scientific domains, we face a fundamental problem when using machine learning to identify proteins from mass spectrometry data: large ground truth datasets mapping inputs to correct outputs are extremely difficult to obtain. Instead, we have access to imperfect hand-coded models crafted by domain experts. In this paper, we apply deep neural networks to an important step of the protein identification problem, the pairing of mass spectra with short sequences of amino acids called peptides. We train our model to differentiate between top scoring results from a state-of-the art classical system and hard-negative second and third place results. Our resulting model is much better at identifying peptides with spectra than the model used to generate its training data. In particular, we achieve a 43% improvement over standard matching methods and a 10% improvement over a combination of the matching method and an industry standard cross-spectra reranking tool. Importantly, in a more difficult experimental regime that reflects current challenges facing biologists, our advantage over the previous state-of-the-art grows to 15% even after reranking. We believe this approach will generalize to other challenging scientific problems.