Properties of average score distributions of SEQUEST: the probability ratio method

High throughput identification of peptides in databases from tandem mass spectrometry data is a key technique in modern proteomics. Common approaches to interpret large scale peptide identification results are based on the statistical analysis of average score distributions, which are constructed from the set of best scores produced by large collections of MS/MS spectra by using searching engines such as SEQUEST. Other approaches calculate individual peptide identification probabilities on the basis of theoretical models or from single-spectrum score distributions constructed by the set of scores produced by each MS/MS spectrum. In this work, we study the mathematical properties of average SEQUEST score distributions by introducing the concept of spectrum quality and expressing these average distributions as compositions of single-spectrum distributions. We predict and demonstrate in the practice that average score distributions are dominated by the quality distribution in the spectra collection, except in the low probability region, where it is possible to predict the dependence of average probability on database size. Our analysis leads to a novel indicator, the probability ratio, which takes optimally into account the statistical information provided by the first and second best scores. The probability ratio is a non-parametric and robust indicator that makes spectra classification according to parameters such as charge state unnecessary and allows a peptide identification performance, on the basis of false discovery rates, that is better than that obtained by other empirical statistical approaches. The probability ratio also compares favorably with statistical probability indicators obtained by the construction of single-spectrum SEQUEST score distributions. These results make the robustness, conceptual simplicity, and ease of automation of the probability ratio algorithm a very attractive alternative to determine peptide identification confidences and error rates in high throughput experiments.     

Profile-based string kernels for remote homology detection and motif extraction

We introduce novel profile-based string kernels for use with support vector machines (SVMs) for the problems of protein classification and remote homology detection. These kernels use probabilistic profiles, such as those produced by the PSI-BLAST algorithm, to define position-dependent mutation neighborhoods along protein sequences for inexact matching of k-length subsequences ("k-mers") in the data. By use of an efficient data structure, the kernels are fast to compute once the profiles have been obtained. For example, the time needed to run PSI-BLAST in order to build the profiles is significantly longer than both the kernel computation time and the SVM training time. We present remote homology detection experiments based on the SCOP database where we show that profile-based string kernels used with SVM classifiers strongly outperform all recently presented supervised SVM methods. We further examine how to incorporate predicted secondary structure information into the profile kernel to obtain a small but significant performance improvement. We also show how we can use the learned SVM classifier to extract "discriminative sequence motifs"--short regions of the original profile that contribute almost all the weight of the SVM classification score--and show that these discriminative motifs correspond to meaningful structural features in the protein data. The use of PSI-BLAST profiles can be seen as a semi-supervised learning technique, since PSI-BLAST leverages unlabeled data from a large sequence database to build more informative profiles. Recently presented "cluster kernels" give general semi-supervised methods for improving SVM protein classification performance. We show that our profile kernel results also outperform cluster kernels while providing much better scalability to large datasets.     

A new algorithm for the evaluation of shotgun peptide sequencing in proteomics: support vector machine classification of peptide MS/MS spectra and SEQUEST scores

Shotgun tandem mass spectrometry-based peptide sequencing using programs such as SEQUEST allows high-throughput identification of peptides, which in turn allows the identification of corresponding proteins. We have applied a machine learning algorithm, called the support vector machine, to discriminate between correctly and incorrectly identified peptides using SEQUEST output. Each peptide was characterized by SEQUEST-calculated features such as delta Cn and Xcorr, measurements such as precursor ion current and mass, and additional calculated parameters such as the fraction of matched MS/MS peaks. The trained SVM classifier performed significantly better than previous cutoff-based methods at separating positive from negative peptides. Positive and negative peptides were more readily distinguished in training set data acquired on a QTOF, compared to an ion trap mass spectrometer. The use of 13 features, including four new parameters, significantly improved the separation between positive and negative peptides. Use of the support vector machine and these additional parameters resulted in a more accurate interpretation of peptide MS/MS spectra and is an important step toward automated interpretation of peptide tandem mass spectrometry data in proteomics.     

The NISTmAb tryptic peptide spectral library for monoclonal antibody characterization

We describe the creation of a mass spectral library composed of all identifiable spectra derived from the tryptic digest of the NISTmAb IgG1κ. The library is a unique reference spectral collection developed from over six million peptide-spectrum matches acquired by liquid chromatography-mass spectrometry (LC-MS) over a wide range of collision energy. Conventional one-dimensional (1D) LC-MS was used for various digestion conditions and 20- and 24-fraction two-dimensional (2D) LC-MS studies permitted in-depth analyses of single digests. Computer methods were developed for automated analysis of LC-MS isotopic clusters to determine the attributes for all ions detected in the 1D and 2D studies. The library contains a selection of over 12,600 high-quality tandem spectra of more than 3,300 peptide ions identified and validated by accurate mass, differential elution pattern, and expected peptide classes in peptide map experiments. These include a variety of biologically modified peptide spectra involving glycosylated, oxidized, deamidated, glycated, and N/C-terminal modified peptides, as well as artifacts. A complete glycation profile was obtained for the NISTmAb with spectra for 58% and 100% of all possible glycation sites in the heavy and light chains, respectively. The site-specific quantification of methionine oxidation in the protein is described. The utility of this reference library is demonstrated by the analysis of a commercial monoclonal antibody (adalimumab, Humira®), where 691 peptide ion spectra are identifiable in the constant regions, accounting for 60% coverage for both heavy and light chains. The NIST reference library platform may be used as a tool for facile identification of the primary sequence and post-translational modifications, as well as the recognition of LC-MS method-induced artifacts for human and recombinant IgG antibodies. Its development also provides a general method for creating comprehensive peptide libraries of individual proteins.     

Cross-scale ecological dynamics and microbial size spectra in marine ecosystems

Evaluating the component features of 'scaling' planktonic size spectra, commonly observed in marine ecosystems, is crucial for understanding the ecological and evolutionary processes from which they emerge. Here, we develop a theoretical framework that describes such spectra in terms of the size distributions of individual species, and test it against actual datasets of microbial size spectra from the Atlantic Ocean. We describe characteristics of size probability distributions of component species that are sufficient to support the observational evidence and infer that, when a power law describes the community size spectrum (thus suggesting critical self-organization of microbial ecosystem structure and function), a related power law links the total number of individuals of a given species to its mean size.     

Rapid identification of dinoflagellates using protein profiling with matrix-assisted laser desorption/ionization mass spectrometry

The occurrence of harmful algal blooms (HABs) or red tides is an important and expanding threat to human health, fishery resources, and the tourism industries. Toxic species post an additional treat of intoxication when consumed either in seafood or directly swallowed. Rapid and accurate identification of the HAB species is critical for minimizing or controlling the damage. We report the use of protein/peptide mass fingerprint profiles obtained with matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF) mass spectrometry (MS) for the identification of dinoflagellates, common causative agents of HABs. The method is simple, fast and reproducible. The peptide mass fingerprint spectral patterns are unique for different dinoflagellate species and are easily distinguishable by visual inspection. In addition to the whole mass spectra, several specific biomarkers were identified from the mass spectra of different species. These biomarker ions and the mass spectral patterns form an unambiguous basis for species discrimination.     

Using dynamic programming to create isotopic distribution maps from mass spectra

MOTIVATION: This article presents a method to identify the isotopic distributions within a mass spectrum using a probabilistic classifier supplemented with dynamic programming. Such a system is needed for a variety of purposes, including generating robust and meaningful features from mass spectra to be used in classification. RESULTS: The primary result of this article is that the dynamic programming approach significantly improves sensitivity, without harming specificity, of a probabilistic classifier for identifying the isotopic distributions. When annotating isotopic distributions where an expert has performed the initial 'peak-picking' (removal of noise peaks), the dynamic programming approach gives a true positive rate of 96% and a false positive rate of 0.0%, whereas the classifier alone has a true positive rate of only 47% when the false positive rate is 0.0%. When annotating isotopic distributions in machine peak-picked spectra, which may contain many noise peaks, the dynamic programming approach gives a true positive rate of only 22.0%, but it still keeps a low false positive rate of 1.0% and still outperforms the classifier alone. It is important to note that all these rates are when we require exact matches with the distributions in annotated spectra; in our evaluation a distribution is considered 'entirely incorrect' if it is missing even one peak or contains even one extraneous peak. We compared to the THRASH and AID-MS systems using a looser requirement: correctly identifying the distribution that contains the mono-isotopic mass. Under this measure, our dynamic programming approach achieves a true positive rate of 82% and a false positive rate of 1%, which again outperforms the classifier alone. The dynamic programming approach ends up being more conservative than THRASH and AID-MS, yielding both fewer true and false peaks, but the F-score of the dynamic programming approach is significantly better than those of THRASH and AID-MS. All results were obtained with 10-fold cross-validation of 99 sections of mass spectra with a total of 214 hand-annotated isotopic distributions. AVAILABILITY: Programs are available via http://www.cs.wisc.edu/~mcilwain/IDM.     

QuickMod: A tool for open modification spectrum library searches

MS2 library spectra are rich in reproducible information about peptide fragmentation patterns compared to theoretical spectra modeled by a sequence search tool. So far, spectrum library searches are mostly applied to detect peptides as they are present in the library. However, they also allow finding modified variants of the library peptides if the search is done with a large precursor mass window and an adapted Spectrum-Spectrum Match (SSM) scoring algorithm. We perform a thorough evaluation on the use of library spectra as opposed to theoretical peptide spectra for the identification of PTMs, analyzing spectra of a well-annotated modification-rich test data set compiled from public data repositories. These initial studies motivate the development of our modification tolerant spectrum library search tool QuickMod, designed to identify modified variants of the peptides listed in the spectrum library without any prior input from the user estimating the modifications present in the sample. We built the search algorithm of QuickMod after carefully testing different SSM similarity scores. The final spectrum scoring scheme uses a support vector machine (SVM) on a selection of scoring features to classify correct and incorrect SSM. After identification of a list of modified peptides at a given False Discovery Rate (FDR), the modifications need to be positioned on the peptide sequence. We present a rapid modification site assignment algorithm and evaluate its positioning accuracy. Finally, we demonstrate that QuickMod performs favorably in terms of speed and identification rate when compared to other software solutions for PTM analysis.     

Factors that contribute to the misidentification of tyrosine nitration by shotgun proteomics

The high selectivity and throughput of tandem mass spectrometry allow for rapid identification and localization of various posttranslational protein modifications from complex mixtures by shotgun approaches. Although sequence database search algorithms provide necessary support to process the potentially enormous quantity of MS/MS spectra generated from large scale tandem mass spectrometry experiments, false positive identifications of peptide modifications may exist even after implementation of stringent identification criteria. In this report, we describe factors that lead to misinterpretation of MS/MS spectra as well as common chemical and experimental artifacts that generate false positives using the proteomics-based identification of tyrosine nitration as an example. In addition to the proposed manual validation criteria, the importance of peptide synthesis and subsequent MS/MS characterization for validation of peptide nitration demonstrated by several examples from earlier publications is also presented.     

The standard protein mix database: a diverse data set to assist in the production of improved Peptide and protein identification software tools

Tandem mass spectrometry (MS/MS) is frequently used in the identification of peptides and proteins. Typical proteomic experiments rely on algorithms such as SEQUEST and MASCOT to compare thousands of tandem mass spectra against the theoretical fragment ion spectra of peptides in a database. The probabilities that these spectrum-to-sequence assignments are correct can be determined by statistical software such as PeptideProphet or through estimations based on reverse or decoy databases. However, many of the software applications that assign probabilities for MS/MS spectra to sequence matches were developed using training data sets from 3D ion-trap mass spectrometers. Given the variety of types of mass spectrometers that have become commercially available over the last 5 years, we sought to generate a data set of reference data covering multiple instrumentation platforms to facilitate both the refinement of existing computational approaches and the development of novel software tools. We analyzed the proteolytic peptides in a mixture of tryptic digests of 18 proteins, named the "ISB standard protein mix", using 8 different mass spectrometers. These include linear and 3D ion traps, two quadrupole time-of-flight platforms (qq-TOF), and two MALDI-TOF-TOF platforms. The resulting data set, which has been named the Standard Protein Mix Database, consists of over 1.1 million spectra in 150+ replicate runs on the mass spectrometers. The data were inspected for quality of separation and searched using SEQUEST. All data, including the native raw instrument and mzXML formats and the PeptideProphet validated peptide assignments, are available at http://regis-web.systemsbiology.net/PublicDatasets/.     

Multi-spectra peptide sequencing and its applications to multistage mass spectrometry

Despite a recent surge of interest in database-independent peptide identifications, accurate de novo peptide sequencing remains an elusive goal. While the recently introduced spectral network approach resulted in accurate peptide sequencing in low-complexity samples, its success depends on the chance of presence of spectra from overlapping peptides. On the other hand, while multistage mass spectrometry (collecting multiple MS 3 spectra from each MS 2 spectrum) can be applied to all spectra in a complex sample, there are currently no software tools for de novo peptide sequencing by multistage mass spectrometry. We describe a rigorous probabilistic framework for analyzing spectra of overlapping peptides and show how to apply it for multistage mass spectrometry. Our software results in both accurate de novo peptide sequencing from multistage mass spectra (despite the inferior quality of MS 3 spectra) and improved interpretation of spectral networks. We further study the problem of de novo peptide sequencing with accurate parent mass (but inaccurate fragment masses), the protocol that may soon become the dominant mode of spectral acquisition. Most existing peptide sequencing algorithms (based on the spectrum graph approach) do not track the accurate parent mass and are thus not equipped for solving this problem. We describe a de novo peptide sequencing algorithm aimed at this experimental protocol and show that it improves the sequencing accuracy on both tandem and multistage mass spectrometry.     

CyclinPred: a SVM-based method for predicting cyclin protein sequences

Functional annotation of protein sequences with low similarity to well characterized protein sequences is a major challenge of computational biology in the post genomic era. The cyclin protein family is once such important family of proteins which consists of sequences with low sequence similarity making discovery of novel cyclins and establishing orthologous relationships amongst the cyclins, a difficult task. The currently identified cyclin motifs and cyclin associated domains do not represent all of the identified and characterized cyclin sequences. We describe a Support Vector Machine (SVM) based classifier, CyclinPred, which can predict cyclin sequences with high efficiency. The SVM classifier was trained with features of selected cyclin and non cyclin protein sequences. The training features of the protein sequences include amino acid composition, dipeptide composition, secondary structure composition and PSI-BLAST generated Position Specific Scoring Matrix (PSSM) profiles. Results obtained from Leave-One-Out cross validation or jackknife test, self consistency and holdout tests prove that the SVM classifier trained with features of PSSM profile was more accurate than the classifiers based on either of the other features alone or hybrids of these features. A cyclin prediction server--CyclinPred has been setup based on SVM model trained with PSSM profiles. CyclinPred prediction results prove that the method may be used as a cyclin prediction tool, complementing conventional cyclin prediction methods.     

Efficient Peak-Labeling Algorithms for Whole-Sample Mass Spectrometry Proteomics

Whole-sample mass spectrometry (MS) proteomics allows for a parallel measurement of hundreds of proteins present in a variety of biospecimens. Unfortunately, the association between MS signals and these proteins is not straightforward. The need to interpret mass spectra demands the development of methods for accurate labeling of ion species in such profiles. To aid this process, we have developed a new peak-labeling procedure for associating protein and peptide labels with peaks. This computational method builds upon characteristics of proteins expected to be in the sample, such as the amino sequence, mass weight, and expected concentration within the sample. A new probabilistic score that incorporates this information is proposed. We evaluate and demonstrate our method's ability to label peaks first on simulated MS spectra and then on MS spectra from human serum with a spiked-in calibration mixture.     

Characterization of the expression ratio noise structure in high-density oligonucleotide arrays

Background

High-density oligonucleotide microarrays provide a powerful tool for assessing differential mRNA expression levels. Characterizing the noise resulting from the enzymatic and hybridization steps, called type I noise, is essential for attributing significance measures to the differential expression scores. We introduce scoring functions for expression ratios, and associated quality measures. Both the PM (Perfect Match) probes and PM-MM differentials (MM is the single MisMatch) are considered as raw intensities. We then characterize the log-ratio noise structure using robust estimates of their intensity dependent variance.

Results

We show the relationships between the obtained ratios and their quality measures. The complementarity of PM and PM-MM methods is emphasized by the probe sets signal to noise measures. Using a large set of replicate experiments, we demonstrate that the noise structure in the log-ratios very closely follows a local log-normal distribution for both the PM and PM-MM cases. Therefore, significance relative to the type I noise can be quantified reliably using the local STD. We discuss the intensity dependence of the STD and show that ratio scores >1.25 are significant in the mid- to high-intensity range.

Conclusions

The ratio noise structure inherent to high-density oligonucleotide arrays can be well described in terms of local log-normal ratio distributions with characteristic intensity dependence. Therefore, robust estimates of the local STD of these distributions provide a simple and powerful way for assessing significance (relative to type I noise) in differential gene expression. This approach will be helpful for improving the reliability of predictions from hybridization experiments in general.     

Algorithms and tools for analysis and management of mass spectrometry data

Mass spectrometry (MS) is a technique that is used for biological studies. It consists in associating a spectrum to a biological sample. A spectrum consists of couples of values (intensity, m/z), where intensity measures the abundance of biomolecules (as proteins) with a mass-to-charge ratio (m/z) present in the originating sample. In proteomics experiments, MS spectra are used to identify pattern expressions in clinical samples that may be responsible of diseases. Recently, to improve the identification of peptides/proteins related to patterns, MS/MS process is used, consisting in performing cascade of mass spectrometric analysis on selected peaks. Latter technique has been demonstrated to improve the identification and quantification of proteins/peptide in samples. Nevertheless, MS analysis deals with a huge amount of data, often affected by noises, thus requiring automatic data management systems. Tools have been developed and most of the time furnished with the instruments allowing: (i) spectra analysis and visualization, (ii) pattern recognition, (iii) protein databases querying, (iv) peptides/proteins quantification and identification. Currently most of the tools supporting such phases need to be optimized to improve the protein (and their functionalities) identification processes. In this article we survey on applications supporting spectrometrists and biologists in obtaining information from biological samples, analyzing available software for different phases. We consider different mass spectrometry techniques, and thus different requirements. We focus on tools for (i) data preprocessing, allowing to prepare results obtained from spectrometers to be analyzed; (ii) spectra analysis, representation and mining, aimed to identify common and/or hidden patterns in spectra sets or in classifying data; (iii) databases querying to identify peptides; and (iv) improving and boosting the identification and quantification of selected peaks. We trace some open problems and report on requirements that represent new challenges for bioinformatics.     

Quality classification of tandem mass spectrometry data

Peptide identification by tandem mass spectrometry is an important tool in proteomic research. Powerful identification programs exist, such as SEQUEST, ProICAT and Mascot, which can relate experimental spectra to the theoretical ones derived from protein databases, thus removing much of the manual input needed in the identification process. However, the time-consuming validation of the peptide identifications is still the bottleneck of many proteomic studies. One way to further streamline this process is to remove those spectra that are unlikely to provide a confident or valid peptide identification, and in this way to reduce the labour from the validation phase. RESULTS: We propose a prefiltering scheme for evaluating the quality of spectra before the database search. The spectra are classified into two classes: spectra which contain valuable information for peptide identification and spectra that are not derived from peptides or contain insufficient information for interpretation. The different spectral features developed for the classification are tested on a real-life material originating from human lymphoblast samples and on a standard mixture of 9 proteins, both labelled with the ICAT-reagent. The results show that the prefiltering scheme efficiently separates the two spectra classes.     

Periodic patterns in distributions of peptide masses

We are investigating the distribution of the number of peptides for given masses, and especially the observation that peptide density reaches a local maximum approximately every 14Da. This wave pattern exists across species (e.g. human or yeast) and enzyme digestion techniques. To analyze this phenomenon we have developed a mathematical method for computing the mass distributions of peptides, and we present both theoretical and empirical evidence that this 14-Da periodicity does not arise from species selection of peptides but from the number- theoretic properties of the masses of amino acid residues. We also describe other, more subtle periodic patterns in the distribution of peptide masses. We also show that these periodic patterns are robust under a variety of conditions, including the addition of amino acid modifications and selection of mass accuracy scale. The method used here is also applicable to any family of sequential molecules, such as linear hydrocarbons, RNA, single- and double-stranded DNA.     

In silico analysis of accurate proteomics, complemented by selective isolation of peptides

Protein identification by mass spectrometry is mainly based on MS/MS spectra and the accuracy of molecular mass determination. However, the high complexity and dynamic ranges for any species of proteomic samples, surpass the separation capacity and detection power of the most advanced multidimensional liquid chromatographs and mass spectrometers. Only a tiny portion of signals is selected for MS/MS experiments and a still considerable number of them do not provide reliable peptide identification. In this article, an in silico analysis for a novel methodology of peptides and proteins identification is described. The approach is based on mass accuracy, isoelectric point (pI), retention time (t(R)) and N-terminal amino acid determination as protein identification criteria regardless of high quality MS/MS spectra. When the methodology was combined with the selective isolation methods, the number of unique peptides and identified proteins increases. Finally, to demonstrate the feasibility of the methodology, an OFFGEL-LC-MS/MS experiment was also implemented. We compared the more reliable peptide identified with MS/MS information, and peptide identified with three experimental features (pI, t(R), molecular mass). Also, two theoretical assumptions from MS/MS identification (selective isolation of peptides and N-terminal amino acid) were analyzed. Our results show that using the information provided by these features and selective isolation methods we could found the 93% of the high confidence protein identified by MS/MS with false-positive rate lower than 5%.     

An approach to the differentiation of leucine and isoleucine residues in EI mass spectra of peptides

Residues of leucine and isoleucine cannot generally be distinguished in the electron impact (EI) generated mass spectra of N-acylated peptide esters. We have obtained the mass spectra of model peptide esters containing leucine or isoleucine in various positions and trifluoroacetyl perdeutero leucine as the N-terminal blocking group. The mass spectra of the peptide derivatives show a pair of peaks as a result of the elimination from the M⁺ ion of neutral fragment of perdeuterated isobutene (M⁺-64) from the leucine side chain of the N-terminal blocking group and isobutene or butene (M⁺-56) from leucine or isoleucine residues of the peptide. The ratios of the intensities of the peaks $\text{M}{}^{\mathrm{+}}\text{-56}\text{M}{}^{\mathrm{+}}\text{-64}$ show considerable variation with the position of leucine or isoleucine in the peptide chain and the length of the peptide, but for peptides which are identical except for the fact that one contains leucine and the other isoleucine in a given position the ratio is always smaller for the isoleucine containing peptide. The differences are sufficient to distinguish the isomeric residues if comparison spectra are available.     

Colander: a probability-based support vector machine algorithm for automatic screening for CID spectra of phosphopeptides prior to database search

We developed a probability-based machine-learning program, Colander, to identify tandem mass spectra that are highly likely to represent phosphopeptides prior to database search. We identified statistically significant diagnostic features of phosphopeptide tandem mass spectra based on ion trap CID MS/MS experiments. Statistics for the features are calculated from 376 validated phosphopeptide spectra and 376 nonphosphopeptide spectra. A probability-based support vector machine (SVM) program, Colander, was then trained on five selected features. Data sets were assembled both from LC/LC-MS/MS analyses of large-scale phosphopeptide enrichments from proteolyzed cells, tissues and synthetic phosphopeptides. These data sets were used to evaluate the capability of Colander to select pS/pT-containing phosphopeptide tandem mass spectra. When applied to unknown tandem mass spectra, Colander can routinely remove 80% of tandem mass spectra while retaining 95% of phosphopeptide tandem mass spectra. The program significantly reduced computational time spent on database search by 60-90%. Furthermore, prefiltering tandem mass spectra representing phosphopeptides can increase the number of phosphopeptide identifications under a predefined false positive rate.