Gradients of Generative Models for Improved Discriminative Analysis of Tandem Mass Spectra
This work addresses a domain-specific problem in proteomics by enhancing discriminative analysis for MS/MS data, representing an incremental improvement over existing hybrid approaches.
The authors tackled the problem of improving peptide identification from tandem mass spectra by leveraging log-likelihood gradients from generative models, resulting in a Fisher kernel that outperformed other methods on evaluated datasets.
Tandem mass spectrometry (MS/MS) is a high-throughput technology used toidentify the proteins in a complex biological sample, such as a drop of blood. A collection of spectra is generated at the output of the process, each spectrum of which is representative of a peptide (protein subsequence) present in the original complex sample. In this work, we leverage the log-likelihood gradients of generative models to improve the identification of such spectra. In particular, we show that the gradient of a recently proposed dynamic Bayesian network (DBN) may be naturally employed by a kernel-based discriminative classifier. The resulting Fisher kernel substantially improves upon recent attempts to combine generative and discriminative models for post-processing analysis, outperforming all other methods on the evaluated datasets. We extend the improved accuracy offered by the Fisher kernel framework to other search algorithms by introducing Theseus, a DBN representing a large number of widely used MS/MS scoring functions. Furthermore, with gradient ascent and max-product inference at hand, we use Theseus to learn model parameters without any supervision.