An improved machine learning protocol for the identification of correct Sequest search results

RaptorX server: a resource for template-based protein structure modeling

Assigning functional properties to a newly discovered protein is a key challenge in modern biology. To this end, computational modeling of the three-dimensional atomic arrangement of the amino acid chain is often crucial in determining the role of the protein in biological processes. We present a community-wide web-based protocol, RaptorX server ( http://raptorx.uchicago.edu ), for automated protein secondary structure prediction, template-based tertiary structure modeling, and probabilistic alignment sampling.Given a target sequence, RaptorX server is able to detect even remotely related template sequences by means of a novel nonlinear context-specific alignment potential and probabilistic consistency algorithm. Using the protocol presented here it is thus possible to obtain high-quality structural models for many target protein sequences when only distantly related protein domains have experimentally solved structures. At present, RaptorX server can perform secondary and tertiary structure prediction of a 200 amino acid target sequence in approximately 30 min.

Improving X!Tandem on peptide identification from mass spectrometry by self-boosted Percolator

A critical component in mass spectrometry (MS)-based proteomics is an accurate protein identification procedure. Database search algorithms commonly generate a list of peptide-spectrum matches (PSMs). The validity of these PSMs is critical for downstream analysis since proteins that are present in the sample are inferred from those PSMs. A variety of postprocessing algorithms have been proposed to validate and filter PSMs. Among them, the most popular ones include a semi-supervised learning (SSL) approach known as Percolator and an empirical modeling approach known as PeptideProphet. However, they are predominantly designed for commercial database search algorithms, i.e., SEQUEST and MASCOT. Therefore, it is highly desirable to extend and optimize those PSM postprocessing algorithms for open source database search algorithms such as X!Tandem. In this paper, we propose a Self-boosted Percolator for postprocessing X!Tandem search results. We find that the SSL algorithm utilized by Percolator depends heavily on the initial ranking of PSMs. Starting with a poor PSM ranking list may cause Percolator to perform suboptimally. By implementing Percolator in a cascade learning manner, we can progressively improve the performance through multiple boost runs, enabling many more PSM identifications without sacrificing false discovery rate (FDR).

Template-based protein structure modeling using the RaptorX web server

A key challenge of modern biology is to uncover the functional role of the protein entities that compose cellular proteomes. To this end, the availability of reliable three-dimensional atomic models of proteins is often crucial. This protocol presents a community-wide web-based method using RaptorX (http://raptorx.uchicago.edu/) for protein secondary structure prediction, template-based tertiary structure modeling, alignment quality assessment and sophisticated probabilistic alignment sampling. RaptorX distinguishes itself from other servers by the quality of the alignment between a target sequence and one or multiple distantly related template proteins (especially those with sparse sequence profiles) and by a novel nonlinear scoring function and a probabilistic-consistency algorithm. Consequently, RaptorX delivers high-quality structural models for many targets with only remote templates. At present, it takes RaptorX ~35 min to finish processing a sequence of 200 amino acids. Since its official release in August 2011, RaptorX has processed ~6,000 sequences submitted by ~1,600 users from around the world.

Search and decoy: the automatic identification of mass spectra

In recent years, the generation and interpretation of MS/MS spectra for the identification of peptides and proteins has matured to a frequently used automatic workflow in Proteomics. Several software solutions for the automated analysis of MS/MS spectra allow for high-throughput/high-performance analyses of complex samples. Related to MS/MS searches, target-decoy approaches have gained more and more popularity: in a "decoy" part of the search database nonexistent sequences mimic real sequences (the "target" sequences). With their help, the number of falsely identified peptides/proteins can be estimated after a search and the resulting protein list can be cut at a specified false discovery rate (FDR). This is an essential prerequisite for all quantitative approaches, as they rely on correct identifications. Especially the label-free approach "spectral counting"-gaining more and more popularity due to low costs and simplicity-depends directly on the correctness of peptide-spectrum matches (PSMs). This work's aim is to describe five popular search engines-especially their general properties regarding protein identification, but also their quantification abilities, if those go beyond spectral counting. By doing so, Proteomics researchers are enabled to compare their features and to choose an appropriate solution for their specific question. Furthermore, the search engines are applied to a spectrum data set generated from a complex sample with a Thermo LTQ Velos OrbiTrap (Thermo Fisher Scientific, Waltham, MA, USA). The results of the search engines are compared, e.g., regarding time requirements, peptides and proteins found, and the search engines' behavior using the decoy approach.