Spectral archives: extending spectral libraries to analyze both identified and unidentified spectra

SpecEncoder: deep metric learning for accurate peptide identification in proteomics

MOTIVATION

Tandem mass spectrometry (MS/MS) is a crucial technology for large-scale proteomic analysis. The protein database search or the spectral library search are commonly used for peptide identification from MS/MS spectra, which, however, may face challenges due to experimental variations between replicated spectra and similar fragmentation patterns among distinct peptides. To address this challenge, we present SpecEncoder, a deep metric learning approach to address these challenges by transforming MS/MS spectra into robust and sensitive embedding vectors in a latent space. The SpecEncoder model can also embed predicted MS/MS spectra of peptides, enabling a hybrid search approach that combines spectral library and protein database searches for peptide identification.

RESULTS

We evaluated SpecEncoder on three large human proteomics datasets, and the results showed a consistent improvement in peptide identification. For spectral library search, SpecEncoder identifies 1%-2% more unique peptides (and PSMs) than SpectraST. For protein database search, it identifies 6%-15% more unique peptides than MSGF+ enhanced by Percolator, Furthermore, SpecEncoder identified 6%-12% additional unique peptides when utilizing a combined library of experimental and predicted spectra. SpecEncoder can also identify more peptides when compared to deep-learning enhanced methods (MSFragger boosted by MSBooster). These results demonstrate SpecEncoder's potential to enhance peptide identification for proteomic data analyses.

AVAILABILITY AND IMPLEMENTATION

The source code and scripts for SpecEncoder and peptide identification are available on GitHub at https://github.com/lkytal/SpecEncoder. Contact: hatang@iu.edu.

A Portable and Reusable Database Infrastructure for Mass Spectrometry, and Its Associated Toolkit (The DIMSpec Project)

Nontargeted analysis (NTA) is a rapidly growing field of techniques that includes the identification of unknown chemical analytes in complex mixtures such as environmental, biological, and food matrices. The use of reference mass spectral databases is a key component of most NTA workflows, providing a high level of confidence for chemical identification when analytical standards are not available, yet effective interlaboratory sharing of research grade spectra remains challenging. The Database Infrastructure for Mass Spectrometry (DIMSpec) project focused on the creation of an open-source toolkit supporting storage and sharing of high-resolution mass spectra with attached sample and methodological metadata. As a demonstration of its utility, the DIMSpec toolkit was used to create a database of curated mass spectra for per- and polyfluoroalkyl substances (PFAS) generated from various sources. While the underlying toolkit is agnostic to analytical targets, this initial release (along with the database schema, mass spectral data, and database tools) should enable PFAS researchers to use these data for their own studies, including the identification of novel PFAS in the environment.

The underappreciated diversity of bile acid modifications

The repertoire of modifications to bile acids and related steroidal lipids by host and microbial metabolism remains incompletely characterized. To address this knowledge gap, we created a reusable resource of tandem mass spectrometry (MS/MS) spectra by filtering 1.2 billion publicly available MS/MS spectra for bile-acid-selective ion patterns. Thousands of modifications are distributed throughout animal and human bodies as well as microbial cultures. We employed this MS/MS library to identify polyamine bile amidates, prevalent in carnivores. They are present in humans, and their levels alter with a diet change from a Mediterranean to a typical American diet. This work highlights the existence of many more bile acid modifications than previously recognized and the value of leveraging public large-scale untargeted metabolomics data to discover metabolites. The availability of a modification-centric bile acid MS/MS library will inform future studies investigating bile acid roles in health and disease.

Fast mass spectrometry search and clustering of untargeted metabolomics data

The throughput of mass spectrometers and the amount of publicly available metabolomics data are growing rapidly, but analysis tools such as molecular networking and Mass Spectrometry Search Tool do not scale to searching and clustering billions of mass spectral data in metabolomics repositories. To address this limitation, we designed MASST+ and Networking+, which can process datasets that are up to three orders of magnitude larger than those processed by state-of-the-art tools.

SPPUSM: An MS/MS spectra merging strategy for improved low-input and single-cell proteome identification

Single and rare cell analysis provides unique insights into the investigation of biological processes and disease progress by resolving the cellular heterogeneity that is masked by bulk measurements. Although many efforts have been made, the techniques used to measure the proteome in trace amounts of samples or in single cells still lag behind those for DNA and RNA due to the inherent non-amplifiable nature of proteins and the sensitivity limitation of current mass spectrometry. Here, we report an MS/MS spectra merging strategy termed SPPUSM (same precursor-produced unidentified spectra merging) for improved low-input and single-cell proteome data analysis. In this method, all the unidentified MS/MS spectra from multiple test files are first extracted. Then, the corresponding MS/MS spectra produced by the same precursor ion from different files are matched according to their precursor mass and retention time (RT) and are merged into one new spectrum. The newly merged spectra with more fragment ions are next searched against the database to increase the MS/MS spectra identification and proteome coverage. Further improvement can be achieved by increasing the number of test files and spectra to be merged. Up to 18.2% improvement in protein identification was achieved for 1 ng HeLa peptides by SPPUSM. Reliability evaluation by the "entrapment database" strategy using merged spectra from human and E. coli revealed a marginal error rate for the proposed method. For application in single cell proteome (SCP) study, identification enhancement of 28%-61% was achieved for proteins for different SCP data. Furthermore, a lower abundance was found for the SPPUSM-identified peptides, indicating its potential for more sensitive low sample input and SCP studies.

Spectroscape enables real-time query and visualization of a spectral archive in proteomics

In proteomics, spectral archives organize the enormous amounts of publicly available peptide tandem mass spectra by similarity, offering opportunities for error correction and novel discoveries. Here we adapt an indexing algorithm developed by Facebook for organizing online multimedia resources to tandem mass spectra and achieve practically instantaneous retrieval and clustering of approximate nearest neighbors in a large spectral archive. An interactive web-based graphical user interface enables the user to view a query spectrum in its clustered neighborhood, which facilitates contextual validation of peptide identifications and exploration of the dark proteome.

Wood Degradation by Fomitiporia mediterranea M. Fischer: Exploring Fungal Adaptation Using Metabolomic Networking

Fomitiporia mediterranea M. Fischer (Fmed) is a white-rot wood-decaying fungus associated with one of the most important and challenging diseases in vineyards: Esca. To relieve microbial degradation, woody plants, including Vitis vinifera, use structural and chemical weapons. Lignin is the most recalcitrant of the wood cell wall structural compounds and contributes to wood durability. Extractives are constitutive or de novo synthesized specialized metabolites that are not covalently bound to wood cell walls and are often associated with antimicrobial properties. Fmed is able to mineralize lignin and detoxify toxic wood extractives, thanks to enzymes such as laccases and peroxidases. Grapevine wood's chemical composition could be involved in Fmed's adaptation to its substrate. This study aimed at deciphering if Fmed uses specific mechanisms to degrade grapevine wood structure and extractives. Three different wood species, grapevine, beech, and oak. were exposed to fungal degradation by two Fmed strains. The well-studied white-rot fungus Trametes versicolor (Tver) was used as a comparison model. A simultaneous degradation pattern was shown for Fmed in the three degraded wood species. Wood mass loss after 7 months for the two fungal species was the highest with low-density oak wood. For the latter wood species, radical differences in initial wood density were observed. No differences between grapevine or beech wood degradation rates were observed after degradation by Fmed or by Tver. Contrary to the Tver secretome, one manganese peroxidase isoform (MnP2l, jgi protein ID 145801) was the most abundant in the Fmed secretome on grapevine wood only. Non-targeted metabolomic analysis was conducted on wood and mycelium samples, using metabolomic networking and public databases (GNPS, MS-DIAL) for metabolite annotations. Chemical differences between non-degraded and degraded woods, and between mycelia grown on different wood species, are discussed. This study highlights Fmed physiological, proteomic and metabolomic traits during wood degradation and thus contributes to a better understanding of its wood degradation mechanisms.

Application of a Novel Hybrid CNN-GNN for Peptide Ion Encoding

Almost all state-of-the-art de novo peptide sequencing algorithms now use machine learning models to encode fragment peaks and hence identify amino acids in mass spectrometry (MS) spectra. Previous work has highlighted how the inherent MS challenges of noise and missing peptide peaks detrimentally affect the performance of these models. In the present research we extracted and evaluated the encoding modules from 3 state-of-the-art de novo peptide sequencing algorithms. We also propose a convolutional neural network-graph neural network machine learning model for encoding peptide ions in tandem MS spectra. We compared the proposed encoding module to those used in the state-of-the-art de novo peptide sequencing algorithms by assessing their ability to identify b-ions and y-ions in MS spectra. This included a comprehensive evaluation in both real and artificial data across various levels of noise and missing peptide peaks. The proposed model performed best across all data sets using two different metrics (area under the receiver operating characteristic curve (AUC) and average precision). The work also highlighted the effect of including additional features such as intensity rank in these encoding modules as well as issues with using the AUC as a metric. This work is of significance to those designing future de novo peptide identification algorithms as it is the first step toward a new approach.

Enhancing untargeted metabolomics using metadata-based source annotation

Human untargeted metabolomics studies annotate only ~10% of molecular features. We introduce reference-data-driven analysis to match metabolomics tandem mass spectrometry (MS/MS) data against metadata-annotated source data as a pseudo-MS/MS reference library. Applying this approach to food source data, we show that it increases MS/MS spectral usage 5.1-fold over conventional structural MS/MS library matches and allows empirical assessment of dietary patterns from untargeted data.

The critical role that spectral libraries play in capturing the metabolomics community knowledge

BACKGROUND

Spectral library searching is currently the most common approach for compound annotation in untargeted metabolomics. Spectral libraries applicable to liquid chromatography mass spectrometry have grown in size over the past decade to include hundreds of thousands to millions of mass spectra and tens of thousands of compounds, forming an essential knowledge base for the interpretation of metabolomics experiments.

AIM OF REVIEW

We describe existing spectral library resources, highlight different strategies for compiling spectral libraries, and discuss quality considerations that should be taken into account when interpreting spectral library searching results. Finally, we describe how spectral libraries are empowering the next generation of machine learning tools in computational metabolomics, and discuss several opportunities for using increasingly accessible large spectral libraries.

KEY SCIENTIFIC CONCEPTS OF REVIEW

This review focuses on the current state of spectral libraries for untargeted LC-MS/MS based metabolomics. We show how the number of entries in publicly accessible spectral libraries has increased more than 60-fold in the past eight years to aid molecular interpretation and we discuss how the role of spectral libraries in untargeted metabolomics will evolve in the near future.

Alignstein: Optimal transport for improved LC-MS retention time alignment

Background

Reproducibility of liquid chromatography separation is limited by retention time drift. As a result, measured signals lack correspondence over replicates of the liquid chromatography–mass spectrometry (LC-MS) experiments. Correction of these errors is named retention time alignment and needs to be performed before further quantitative analysis. Despite the availability of numerous alignment algorithms, their accuracy is limited (e.g., for retention time drift that swaps analytes’ elution order).

Results

We present the Alignstein, an algorithm for LC-MS retention time alignment. It correctly finds correspondence even for swapped signals. To achieve this, we implemented the generalization of the Wasserstein distance to compare multidimensional features without any reduction of the information or dimension of the analyzed data. Moreover, Alignstein by design requires neither a reference sample nor prior signal identification. We validate the algorithm on publicly available benchmark datasets obtaining competitive results. Finally, we show that it can detect the information contained in the tandem mass spectrum by the spatial properties of chromatograms.

Conclusions

We show that the use of optimal transport effectively overcomes the limitations of existing algorithms for statistical analysis of mass spectrometry datasets. The algorithm’s source code is available at https://github.com/grzsko/Alignstein.

Higher-Energy Collisional Dissociation Mass Spectral Networks for the Rapid, Semi-automated Characterization of Known and Unknown Ribonucleoside Modifications

Higher-energy collisional dissociation (HCD) of modified ribonucleosides generates characteristic and highly reproducible nucleoside-specific tandem mass spectra (MS/MS). Here, we demonstrate the capability of HCD spectra in combination with spectral matching for the semi-automated characterization of ribonucleosides. This process involved the generation of an HCD spectral library and the establishment of a mass spectral network for rapid detection with high sensitivity and specificity in a retention time-independent fashion. Systematic spectral matching analysis of the MS/MS spectra of tRNA hydrolysates from different organisms has helped us to uncover evidence for the existence of novel ribonucleoside modifications such as s²Cm and OHyW-14. Such an untargeted label-free approach has the potential to be integrated with other methods, including those that use isotope labeling, to simplify the characterization of unknown modified ribonucleosides. These findings suggest the compilation of a universal spectral network, for the characterization of known and unknown ribonucleosides, could accelerate discoveries in the epitranscriptome.

A Comprehensive Evaluation of Consensus Spectrum Generation Methods in Proteomics

Spectrum clustering is a powerful strategy to minimize redundant mass spectra by grouping them based on similarity, with the aim of forming groups of mass spectra from the same repeatedly measured analytes. Each such group of near-identical spectra can be represented by its so-called consensus spectrum for downstream processing. Although several algorithms for spectrum clustering have been adequately benchmarked and tested, the influence of the consensus spectrum generation step is rarely evaluated. Here, we present an implementation and benchmark of common consensus spectrum algorithms, including spectrum averaging, spectrum binning, the most similar spectrum, and the best-identified spectrum. We have analyzed diverse public data sets using two different clustering algorithms (spectra-cluster and MaRaCluster) to evaluate how the consensus spectrum generation procedure influences downstream peptide identification. The BEST and BIN methods were found the most reliable methods for consensus spectrum generation, including for data sets with post-translational modifications (PTM) such as phosphorylation. All source code and data of the present study are freely available on GitHub at https://github.com/statisticalbiotechnology/representative-spectra-benchmark.

SIMSI-Transfer: Software-assisted reduction of missing values in phosphoproteomic and proteomic isobaric labeling data using tandem mass spectrum clustering

Isobaric stable isotope labeling techniques such as tandem mass tags (TMTs) have become popular in proteomics because they enable the relative quantification of proteins with high precision from up to 18 samples in a single experiment. While missing values in peptide quantification are rare in a single TMT experiment, they rapidly increase when combining multiple TMT experiments. As the field moves toward analyzing ever higher numbers of samples, tools that reduce missing values also become more important for analyzing TMT datasets. To this end, we developed SIMSI-Transfer (Similarity-based Isobaric Mass Spectra 2 [MS2] Identification Transfer), a software tool that extends our previously developed software MaRaCluster (© Matthew The) by clustering similar tandem MS2 from multiple TMT experiments. SIMSI-Transfer is based on the assumption that similarity-clustered MS2 spectra represent the same peptide. Therefore, peptide identifications made by database searching in one TMT batch can be transferred to another TMT batch in which the same peptide was fragmented but not identified. To assess the validity of this approach, we tested SIMSI-Transfer on masked search engine identification results and recovered >80% of the masked identifications while controlling errors in the transfer procedure to below 1% false discovery rate. Applying SIMSI-Transfer to six published full proteome and phosphoproteome datasets from the Clinical Proteomic Tumor Analysis Consortium led to an increase of 26 to 45% of identified MS2 spectra with TMT quantifications. This significantly decreased the number of missing values across batches and, in turn, increased the number of peptides and proteins identified in all TMT batches by 43 to 56% and 13 to 16%, respectively.

•

Spectrum clustering enables peptide identification transfer between LC–MS/MS runs.

•

The SIMSI pipeline supports processing full proteome and phosphoproteome data.

•

SIMSI increases the number of quantifiable PSMs by 26 to 45%.

•

SIMSI reduces missing values in multibatch TMT labeling experiments by up to 21%.

Optimization of LC-MS2 Data Acquisition Parameters for Molecular Networking Applied to Marine Natural Products

Since the introduction of the online open-source GNPS, molecular networking has quickly become a widely applied tool in the field of natural products chemistry, with applications from dereplication, genome mining, metabolomics, and visualization of chemical space. Studies have shown that data dependent acquisition (DDA) parameters affect molecular network topology but are limited in the number of parameters studied. With an aim to optimize LC-MS² parameters for integrating GNPS-based molecular networking into our marine natural products workflow, a design of experiment (DOE) was used to screen the significance of the effect that eleven parameters have on both Classical Molecular Networking workflow (CLMN) and the new Feature-Based Molecular Networking workflow (FBMN). Our results indicate that four parameters (concentration, run duration, collision energy and number of precursors per cycle) are the most significant data acquisition parameters affecting the network topology. While concentration and the LC duration were found to be the two most important factors to optimize for CLMN, the number of precursors per cycle and collision energy were also very important factors to optimize for FBMN.

The impact of noise and missing fragmentation cleavages on de novo peptide identification algorithms

•

Most correct de novo peptides have ⩽1 missing fragmentation cleavages.

•

DeepNovo outperforms Novor for peptide accuracy for both data types.

•

Novor excels at amino acid recall when many fragmentation cleavages are missing.

•

Deep learning allows DeepNovo to predict amino acids without adjacent peaks.

Proteomics aims to characterise system-wide protein expression and typically relies on mass-spectrometry and peptide fragmentation, followed by a database search for protein identification. It has wide ranging applications from clinical to environmental settings and virtually impacts on every area of biology. In that context, de novo peptide sequencing is becoming increasingly popular. Historically its performance lagged behind database search methods but with the integration of machine learning, this field of research is gaining momentum. To enable de novo peptide sequencing to realise its full potential, it is critical to explore the mass spectrometry data underpinning peptide identification. In this research we investigate the characteristics of tandem mass spectra using 8 published datasets. We then evaluate two state of the art de novo peptide sequencing algorithms, Novor and DeepNovo, with a particular focus on their performance with regard to missing fragmentation cleavage sites and noise. DeepNovo was found to perform better than Novor overall. However, Novor recalled more correct amino acids when 6 or more cleavage sites were missing. Furthermore, less than 11% of each algorithms’ correct peptide predictions emanate from data with more than one missing cleavage site, highlighting the issues missing cleavages pose. We further investigate how the algorithms manage to correctly identify peptides with many of these missing fragmentation cleavages. We show how noise negatively impacts the performance of both algorithms, when high intensity peaks are considered. Finally, we provide recommendations regarding further algorithms’ improvements and offer potential avenues to overcome current inherent data limitations.

Data Processing of Product Ion Spectra: Quality Improvement by Averaging Multiple Similar Spectra of Small Molecules

In metabolomics studies using high-resolution mass spectrometry (MS), a set of product ion spectra is comprehensively acquired from observed ions using the data-dependent acquisition (DDA) mode of various tandem MS. However, especially for low-intensity signals, it is sometimes difficult to distinguish artifact signals from true fragment ions derived from a precursor ion. Inadequate precision in the measured m/z value is also one of the bottlenecks to narrowing down the candidate compositional formula. In this study, we report that averaging multiple product ion spectra can improve m/z precision as well as the reliability of fragment ions that are observed in such spectra. A graph-based method was applied to cluster a set of similar spectra from multiple DDA data files resulting in creating an averaged product-ion spectrum. The error levels for the m/z values declined following the central limit theorem, which allowed us to reduce the number of candidate compositional formulas. The improved reliability and precision of the averaged spectra will contribute to a more efficient annotation of product ion spectral data.

Combining multiple acquisition modes and computational data annotation for structural characterization in traditional Chinese medicine: Miao Nationality medicine Qijiao Shengbai Capsule as a case study

Qijiao Shengbai Capsule (QSC) is a reputable Miao Nationality medicine used for treating leukopenia, but its chemical composition has not yet been elucidated. We herein present a strategy, by integrating multiple data acquisition, computational data annotation and processing methods to visualize and identify the complicated constituents in QSC based on ultra-high-performance liquid chromatography coupled with traveling wave ion mobility quadrupole time-of-flight mass spectrometry (UPLC-TWIMS-QTOF-MS). The multiple data acquisition modes, including data-independent mass spectrometry^Energy (MS^E), data-independent high-definition mass spectrometry^Energy (HDMS^E), and fast data-dependent acquisition (fast-DDA), in both positive and negative ion modes, were conducted on a Waters-SYNAPT G2-Si mass spectrometer with an ESI source. An in-house library built by the UNIFI platform could efficiently process the peak annotation of known compounds, whilst different structural types were clustered in the molecular networks for the analogous classification and structural annotation of the unknown ones. Neutral loss, diagnostic ions, feature fragmentation behaviors, and community curation of mass spectrometry data of known compounds helped exploit those similar neighboring nodes of unknown compounds. Moreover, by combination of the predicted CCS values from CCS platform with the experimental CCS values from HDMS^E, as well as diagnostic fragment ions, isomer compounds were annotated. By integrating reference compound comparison, a total of 202 constituents, including 94 flavonoids, 12 saponins, 30 phthalides, 38 organic acids, 3 amino acids, 7 alkaloids and 18 others, were unambiguously characterized or tentatively identified in QSC. Among them, 5 potential new compounds were detected and 12 pairs of isomers were comprehensively distinguished. Conclusively, the established multiple acquisition modes, computational data processing and analysis strategy proved to be useful for the in-depth structural identification of QSC.

Qijiao Shengbai Capsule (QSC) is a reputable Miao Nationality medicine used for treating leukopenia, but its chemical composition has not yet been elucidated.

ClusterSheep: A Graphics Processing Unit-Accelerated Software Tool for Large-Scale Clustering of Tandem Mass Spectra from Shotgun Proteomics

Modern shotgun proteomics experiments generate gigabytes of spectra every hour, only a fraction of which were utilized to form biological conclusions. Instead of being stored as flat files in public data repositories, this large amount of data can be better organized to facilitate data reuse. Clustering these spectra by similarity can be helpful in building high-quality spectral libraries, correcting identification errors, and highlighting frequently observed but unidentified spectra. However, large-scale clustering is time-consuming. Here, we present ClusterSheep, a method utilizing Graphics Processing Units (GPUs) to accelerate the process. Unlike previously proposed algorithms for this purpose, our method performs true pairwise comparison of all spectra within a precursor mass-to-charge ratio tolerance, thereby preserving the full cluster structures. ClusterSheep was benchmarked against previously reported clustering tools, MS-Cluster, MaRaCluster, and msCRUSH. The software tool also functions as an interactive visualization tool with a persistent state, enabling the user to explore the resulting clusters visually and retrieve the clustering results as desired.

Expanding Urinary Metabolite Annotation through Integrated Mass Spectral Similarity Networking

The urine metabolome constitutes a rich source of functional information reflecting physiological states that are influenced by distinct conditions and biological stresses, such as responses to drug treatments or disease manifestations. Although global liquid chromatography-mass spectrometry (MS) profiling provides the most comprehensive measurement of metabolites in complex biological samples, annotation remains a challenge, and computational approaches are necessary to translate the molecular composition into biological knowledge. Here, we investigated the use of tandem MS-based enhanced molecular networks (MolNetEnhancer) to improve the metabolite annotation of urine extracts. The samples (n = 10) were analyzed by hydrophilic interaction chromatography-quadrupole time-of-flight mass spectrometry in both electrospray ionization (ESI) modes. Consistent with other common data preprocessing software, the use of Progenesis QI led to the annotation of up to 20 metabolites based on MS2 library searches, showing a high fragmentation score (cosine similarity ≥ 0.7), that is, ∼2% of mass features containing MS2 spectra. Molecular networking based on library matching resulted in the annotation of up to 62 urinary compounds. Using a combination of unsupervised substructure discovery (MS2LDA), the in silico tool network annotation propagation (NAP), and ClassyFire chemical ontology, embedded in a multilayered molecular network by MolNetEnhancer, we were able to expand the chemical characterization to ∼50% of the data set. The integrative approach led to the annotation of 275 compounds at the metabolomics standards initiative (MSI) confidence level 2, as well as 459 and 578 urinary metabolites (MSI level 3) in both negative and positive ESI modes, respectively. The exhaustive MS2-based annotation outperformed similar studies applied to larger cohorts while offering the discovery of metabolites not identified by the MS2 library search. This is the first work that effectively integrates orthogonal annotation methods and MS2-based fragmentation studies to improve metabolite annotation in urine samples.

Integrating genomics and metabolomics for scalable non-ribosomal peptide discovery

Non-Ribosomal Peptides (NRPs) represent a biomedically important class of natural products that include a multitude of antibiotics and other clinically used drugs. NRPs are not directly encoded in the genome but are instead produced by metabolic pathways encoded by biosynthetic gene clusters (BGCs). Since the existing genome mining tools predict many putative NRPs synthesized by a given BGC, it remains unclear which of these putative NRPs are correct and how to identify post-assembly modifications of amino acids in these NRPs in a blind mode, without knowing which modifications exist in the sample. To address this challenge, here we report NRPminer, a modification-tolerant tool for NRP discovery from large (meta)genomic and mass spectrometry datasets. We show that NRPminer is able to identify many NRPs from different environments, including four previously unreported NRP families from soil-associated microbes and NRPs from human microbiota. Furthermore, in this work we demonstrate the anti-parasitic activities and the structure of two of these NRP families using direct bioactivity screening and nuclear magnetic resonance spectrometry, illustrating the power of NRPminer for discovering bioactive NRPs.

Anaerobic gut fungi are an untapped reservoir of natural products

Anaerobic gut fungi are important members of the gut microbiome of herbivores, yet they exist in small numbers relative to bacteria. Here, we show that these early-branching fungi produce a wealth of secondary metabolites (natural products) that may act to regulate the gut microbiome. We use an integrated 'omics'-based approach to classify the biosynthetic genes predicted from fungal genomes, determine transcriptionally active genes, and verify the presence of their enzymatic products. Our analysis reveals that anaerobic gut fungi are an untapped reservoir of bioactive compounds that could be harnessed for biotechnology.

Anaerobic fungi (class Neocallimastigomycetes) thrive as low-abundance members of the herbivore digestive tract. The genomes of anaerobic gut fungi are poorly characterized and have not been extensively mined for the biosynthetic enzymes of natural products such as antibiotics. Here, we investigate the potential of anaerobic gut fungi to synthesize natural products that could regulate membership within the gut microbiome. Complementary 'omics' approaches were combined to catalog the natural products of anaerobic gut fungi from four different representative species: Anaeromyces robustus (A. robustus), Caecomyces churrovis (C. churrovis), Neocallimastix californiae (N. californiae), and Piromyces finnis (P. finnis). In total, 146 genes were identified that encode biosynthetic enzymes for diverse types of natural products, including nonribosomal peptide synthetases and polyketide synthases. In addition, N. californiae and C. churrovis genomes encoded seven putative bacteriocins, a class of antimicrobial peptides typically produced by bacteria. During standard laboratory growth on plant biomass or soluble substrates, 26% of total core biosynthetic genes in all four strains were transcribed. Across all four fungal strains, 30% of total biosynthetic gene products were detected via proteomics when grown on cellobiose. Liquid chromatography-tandem mass spectrometry (LC-MS/MS) characterization of fungal supernatants detected 72 likely natural products from A. robustus alone. A compound produced by all four strains of anaerobic fungi was putatively identified as the polyketide-related styrylpyrone baumin. Molecular networking quantified similarities between tandem mass spectrometry (MS/MS) spectra among these fungi, enabling three groups of natural products to be identified that are unique to anaerobic fungi. Overall, these results support the finding that anaerobic gut fungi synthesize natural products, which could be harnessed as a source of antimicrobials, therapeutics, and other bioactive compounds.

CHICKN: extraction of peptide chromatographic elution profiles from large scale mass spectrometry data by means of Wasserstein compressive hierarchical cluster analysis

Background

The clustering of data produced by liquid chromatography coupled to mass spectrometry analyses (LC-MS data) has recently gained interest to extract meaningful chemical or biological patterns. However, recent instrumental pipelines deliver data which size, dimensionality and expected number of clusters are too large to be processed by classical machine learning algorithms, so that most of the state-of-the-art relies on single pass linkage-based algorithms.

Results

We propose a clustering algorithm that solves the powerful but computationally demanding kernel k-means objective function in a scalable way. As a result, it can process LC-MS data in an acceptable time on a multicore machine. To do so, we combine three essential features: a compressive data representation, Nyström approximation and a hierarchical strategy. In addition, we propose new kernels based on optimal transport, which interprets as intuitive similarity measures between chromatographic elution profiles.

Conclusions

Our method, referred to as CHICKN, is evaluated on proteomics data produced in our lab, as well as on benchmark data coming from the literature. From a computational viewpoint, it is particularly efficient on raw LC-MS data. From a data analysis viewpoint, it provides clusters which differ from those resulting from state-of-the-art methods, while achieving similar performances. This highlights the complementarity of differently principle algorithms to extract the best from complex LC-MS data.

Visible Post-Data Analysis Protocol for Natural Mycotoxin Production

Fungal natural products are routinely analyzed using target detection protocols by comparing to commercial standards. However, discovery of new products suffers from a lack of high-throughput analytical techniques. Post-data process techniques have become popular tools for natural product confirmations and mycotoxin family analysis. In this work, a visible post-data process procedure with MZmine, GNPS, and Xcalibur was used for efficient analysis of high-resolution mass spectrometry. Conjugated products were screened with an optimized diagnostic fragmentation filtering module in MZmine and further confirmed with Xcalibur by comparing to unconjugated commercial standards. MS/MS spectral data were processed and used to establish a feature based on a molecular networking map in GNPS (Global Natural Products Social Molecular Networking; https://gnps.ucsd.edu), for visualization of fungal natural product families. The results demonstrate the potential of combining MZmine-, GNPS-, and Xcalibur-based methods for visible analysis of fungal natural products.

Focus on the spectra that matter by clustering of quantification data in shotgun proteomics

In shotgun proteomics, the analysis of label-free quantification experiments is typically limited by the identification rate and the noise level in the quantitative data. This generally causes a low sensitivity in differential expression analysis. Here, we propose a quantification-first approach for peptides that reverses the classical identification-first workflow, thereby preventing valuable information from being discarded in the identification stage. Specifically, we introduce a method, Quandenser, that applies unsupervised clustering on both MS1 and MS2 level to summarize all analytes of interest without assigning identities. This reduces search time due to the data reduction. We can now employ open modification and de novo searches to identify analytes of interest that would have gone unnoticed in traditional pipelines. Quandenser+Triqler outperforms the state-of-the-art method MaxQuant+Perseus, consistently reporting more differentially abundant proteins for all tested datasets. Software is available for all major operating systems at https://github.com/statisticalbiotechnology/quandenser, under Apache 2.0 license.

Reproducible molecular networking of untargeted mass spectrometry data using GNPS

Global Natural Product Social Molecular Networking (GNPS) is an interactive online small molecule-focused tandem mass spectrometry (MS2) data curation and analysis infrastructure. It is intended to provide as much chemical insight as possible into an untargeted MS2 dataset and to connect this chemical insight to the user's underlying biological questions. This can be performed within one liquid chromatography (LC)-MS2 experiment or at the repository scale. GNPS-MassIVE is a public data repository for untargeted MS2 data with sample information (metadata) and annotated MS2 spectra. These publicly accessible data can be annotated and updated with the GNPS infrastructure keeping a continuous record of all changes. This knowledge is disseminated across all public data; it is a living dataset. Molecular networking-one of the main analysis tools used within the GNPS platform-creates a structured data table that reflects the molecular diversity captured in tandem mass spectrometry experiments by computing the relationships of the MS2 spectra as spectral similarity. This protocol provides step-by-step instructions for creating reproducible, high-quality molecular networks. For training purposes, the reader is led through a 90- to 120-min procedure that starts by recalling an example public dataset and its sample information and proceeds to creating and interpreting a molecular network. Each data analysis job can be shared or cloned to disseminate the knowledge gained, thus propagating information that can lead to the discovery of molecules, metabolic pathways, and ecosystem/community interactions.

Anatomy and evolution of database search engines-a central component of mass spectrometry based proteomic workflows

Sequence database search engines are bioinformatics algorithms that identify peptides from tandem mass spectra using a reference protein sequence database. Two decades of development, notably driven by advances in mass spectrometry, have provided scientists with more than 30 published search engines, each with its own properties. In this review, we present the common paradigm behind the different implementations, and its limitations for modern mass spectrometry datasets. We also detail how the search engines attempt to alleviate these limitations, and provide an overview of the different software frameworks available to the researcher. Finally, we highlight alternative approaches for the identification of proteomic mass spectrometry datasets, either as a replacement for, or as a complement to, sequence database search engines.

A Perspective on the Confident Comparison of Glycoprotein Site-Specific Glycosylation in Sample Cohorts

Protein glycosylation, resulting from glycosyl transferase reactions under complex control in the secretory pathway, consists of a distribution of related glycoforms at each glycosylation site. Because the biosynthetic substrate concentration and transport rates depend on architecture and other aspects of cellular phenotypes, site-specific glycosylation cannot be predicted accurately from genomic, transcriptomic, or proteomic information. Rather, it is necessary to quantify glycosylation at each protein site and how this changes among a sample cohort to provide information about disease mechanisms. At present, mature mass spectrometry-based methods allow for qualitative assignment of the glycan composition and glycosylation site of singly glycosylated proteolytic peptides. To make such quantitative comparisons, it is necessary to sample the glycosylation distribution with sufficient coverage and accuracy for confident assessment of the glycosylation changes that occur in the biological cohort. In this Perspective, we discuss the unmet needs for mass spectrometry acquisition methods and bioinformatics for the confident comparison of protein site-specific glycosylation among sample cohorts.

De Novo Peptide Sequencing Reveals Many Cyclopeptides in the Human Gut and Other Environments

Cyclic and branch cyclic peptides (cyclopeptides) represent a class of bioactive natural products that include many antibiotics and anti-tumor compounds. Despite the recent advances in metabolomics analysis, still little is known about the cyclopeptides in the human gut and their possible interactions due to a lack of computational analysis pipelines that are applicable to such compounds. Here, we introduce CycloNovo, an algorithm for automated de novo cyclopeptide analysis and sequencing that employs de Bruijn graphs, the workhorse of DNA sequencing algorithms, to identify cyclopeptides in spectral datasets. CycloNovo reconstructed 32 previously unreported cyclopeptides (to the best of our knowledge) in the human gut and reported over a hundred cyclopeptides in other environments represented by various spectra on Global Natural Products Social Molecular Network (GNPS). https://github.com/bbehsaz/cyclonovo.

Combination of Isotope Labeling and Molecular Networking of Tandem Mass Spectrometry Data To Reveal 69 Unknown Metabolites Produced by Penicillium nordicum

The secondary metabolome of Penicillium nordicum is poorly documented despite its frequent detection on contaminated food and its capacity to produce toxic metabolites such as ochratoxin A. To characterize metabolites produced by this fungi, we combined a full stable isotopes labeling with the dereplication of tandem mass spectrometry (MS/MS) data by molecular networking. First, the untargeted metabolomic analysis by high-resolution mass spectrometry of a double stable isotope labeling of P. nordicum enabled the specific detection of its metabolites and the unambiguous determination of their elemental composition. Analyses showed that infection of substrate by P. nordicum lead to the production of at least 92 metabolites and that 69 of them were completely unknown. Then, curated molecular networks of MS/MS data were generated with GNPS and MetGem, specifically on the features of interest, which allowed highlighting 13 fungisporin-related metabolites that had not previously been identified in this fungus and 8 that had never been observed in any fungus. The structures of the unknown compounds, namely, a native fungisporin and seven linear peptides, were characterized by tandem mass spectrometry experiments. The analysis of P. nordicum growing on its natural substrates, i.e. pork ham, turkey ham, and cheese, demonstrated that 10 of the known fungisporin-related metabolites and three of the new metabolites were also synthesized. Thus, the curation of data for molecular networking using a specific detection of metabolites of interest with stable isotopes labeling allowed the discovery of new metabolites produced by the food contaminant P. nordicum.

Mass Spectral Similarity Networking and Gas-Phase Fragmentation Reactions in the Structural Analysis of Flavonoid Glycoconjugates

Flavonoids represent an important class of natural products with a central role in plant physiology and human health. Their accurate annotation using untargeted mass spectrometry analysis still relies on differentiating similar chemical scaffolds through spectral matching to reference library spectra. In this work, we combined molecular network analysis with rules for fragment reactions and chemotaxonomy to enhance the annotation of similar flavonoid glyconjugates. Molecular network topology progressively propagated the flavonoid chemical functionalization according to collision-induced dissociation (CID) reactions, as the following chemical attributes: aglycone nature, saccharide type and number, and presence of methoxy substituents. This structure-based distribution across the spectral networks revealed the chemical composition of flavonoids across intra- and interspecies and guided the putatively assignment of 64 isomers and isobars in the Chrysobalanaceae plant species, most of which are not accurately annotated by automated untargeted MS² matching. These proof of concept results demonstrate how molecular networking progressively grouped structurally related molecules according to their product ion scans, abundances, and ratios. The approach can be extrapolated to other classes of metabolites sharing similar structures and diagnostic fragments from tandem mass spectrometry.

The impact of skin care products on skin chemistry and microbiome dynamics

BACKGROUND

Use of skin personal care products on a regular basis is nearly ubiquitous, but their effects on molecular and microbial diversity of the skin are unknown. We evaluated the impact of four beauty products (a facial lotion, a moisturizer, a foot powder, and a deodorant) on 11 volunteers over 9 weeks.

RESULTS

Mass spectrometry and 16S rRNA inventories of the skin revealed decreases in chemical as well as in bacterial and archaeal diversity on halting deodorant use. Specific compounds from beauty products used before the study remain detectable with half-lives of 0.5-1.9 weeks. The deodorant and foot powder increased molecular, bacterial, and archaeal diversity, while arm and face lotions had little effect on bacterial and archaeal but increased chemical diversity. Personal care product effects last for weeks and produce highly individualized responses, including alterations in steroid and pheromone levels and in bacterial and archaeal ecosystem structure and dynamics.

CONCLUSIONS

These findings may lead to next-generation precision beauty products and therapies for skin disorders.

Spectral Clustering Improves Label-Free Quantification of Low-Abundant Proteins

Label-free quantification has become a common-practice in many mass spectrometry-based proteomics experiments. In recent years, we and others have shown that spectral clustering can considerably improve the analysis of (primarily large-scale) proteomics data sets. Here we show that spectral clustering can be used to infer additional peptide-spectrum matches and improve the quality of label-free quantitative proteomics data in data sets also containing only tens of MS runs. We analyzed four well-known public benchmark data sets that represent different experimental settings using spectral counting and peak intensity based label-free quantification. In both approaches, the additionally inferred peptide-spectrum matches through our spectra-cluster algorithm improved the detectability of low abundant proteins while increasing the accuracy of the derived quantitative data, without increasing the data sets' noise. Additionally, we developed a Proteome Discoverer node for our spectra-cluster algorithm which allows anyone to rebuild our proposed pipeline using the free version of Proteome Discoverer.

Expanding the Use of Spectral Libraries in Proteomics

The 2017 Dagstuhl Seminar on Computational Proteomics provided an opportunity for a broad discussion on the current state and future directions of the generation and use of peptide tandem mass spectrometry spectral libraries. Their use in proteomics is growing slowly, but there are multiple challenges in the field that must be addressed to further increase the adoption of spectral libraries and related techniques. The primary bottlenecks are the paucity of high quality and comprehensive libraries and the general difficulty of adopting spectral library searching into existing workflows. There are several existing spectral library formats, but none captures a satisfactory level of metadata; therefore, a logical next improvement is to design a more advanced, Proteomics Standards Initiative-approved spectral library format that can encode all of the desired metadata. The group discussed a series of metadata requirements organized into three designations of completeness or quality, tentatively dubbed bronze, silver, and gold. The metadata can be organized at four different levels of granularity: at the collection (library) level, at the individual entry (peptide ion) level, at the peak (fragment ion) level, and at the peak annotation level. Strategies for encoding mass modifications in a consistent manner and the requirement for encoding high-quality and commonly seen but as-yet-unidentified spectra were discussed. The group also discussed related topics, including strategies for comparing two spectra, techniques for generating representative spectra for a library, approaches for selection of optimal signature ions for targeted workflows, and issues surrounding the merging of two or more libraries into one. We present here a review of this field and the challenges that the community must address in order to accelerate the adoption of spectral libraries in routine analysis of proteomics datasets.

Identification of small molecules using accurate mass MS/MS search

Tandem mass spectral library search (MS/MS) is the fastest way to correctly annotate MS/MS spectra from screening small molecules in fields such as environmental analysis, drug screening, lipid analysis, and metabolomics. The confidence in MS/MS-based annotation of chemical structures is impacted by instrumental settings and requirements, data acquisition modes including data-dependent and data-independent methods, library scoring algorithms, as well as post-curation steps. We critically discuss parameters that influence search results, such as mass accuracy, precursor ion isolation width, intensity thresholds, centroiding algorithms, and acquisition speed. A range of publicly and commercially available MS/MS databases such as NIST, MassBank, MoNA, LipidBlast, Wiley MSforID, and METLIN are surveyed. In addition, software tools including NIST MS Search, MS-DIAL, Mass Frontier, SmileMS, Mass++, and XCMS2 to perform fast MS/MS search are discussed. MS/MS scoring algorithms and challenges during compound annotation are reviewed. Advanced methods such as the in silico generation of tandem mass spectra using quantum chemistry and machine learning methods are covered. Community efforts for curation and sharing of tandem mass spectra that will allow for faster distribution of scientific discoveries are discussed.

Response to "Comparison and Evaluation of Clustering Algorithms for Tandem Mass Spectra"

In the recent benchmarking article entitled "Comparison and Evaluation of Clustering Algorithms for Tandem Mass Spectra", Rieder et al. compared several different approaches to cluster MS/MS spectra. While we certainly recognize the value of the manuscript, here, we report some shortcomings detected in the original analyses. For most analyses, the authors clustered only single MS/MS runs. In one of the reported analyses, three MS/MS runs were processed together, which already led to computational performance issues in many of the tested approaches. This fact highlights the difficulties of using many of the tested algorithms on the nowadays produced average proteomics data sets. Second, the authors only processed identified spectra when merging MS runs. Thereby, all unidentified spectra that are of lower quality were already removed from the data set and could not influence the clustering results. Next, we found that the authors did not analyze the effect of chimeric spectra on the clustering results. In our analysis, we found that 3% of the spectra in the used data sets were chimeric, and this had marked effects on the behavior of the different clustering algorithms tested. Finally, the authors' choice to evaluate the MS-Cluster and spectra-cluster algorithms using a precursor tolerance of 5 Da for high-resolution Orbitrap data only was, in our opinion, not adequate to assess the performance of MS/MS clustering approaches.

Tandem mass spectral libraries of peptides and their roles in proteomics research

Proteomics is a rapidly maturing field aimed at the high-throughput identification and quantification of all proteins in a biological system. The cornerstone of proteomic technology is tandem mass spectrometry of peptides resulting from the digestion of protein mixtures. The fragmentation pattern of each peptide ion is captured in its tandem mass spectrum, which enables its identification and acts as a fingerprint for the peptide. Spectral libraries are simply searchable collections of these fingerprints, which have taken on an increasingly prominent role in proteomic data analysis. This review describes the historical development of spectral libraries in proteomics, details the computational procedures behind library building and searching, surveys the current applications of spectral libraries, and discusses the outstanding challenges. © 2016 Wiley Periodicals, Inc. Mass Spec Rev 36:634-648, 2017.

Imaging mass spectrometry and MS/MS molecular networking reveals chemical interactions among cuticular bacteria and pathogenic fungi associated with fungus-growing ants

The fungus-growing ant-microbe symbiosis is an ideal system to study chemistry-based microbial interactions due to the wealth of microbial interactions described, and the lack of information on the molecules involved therein. In this study, we employed a combination of MALDI imaging mass spectrometry (MALDI-IMS) and MS/MS molecular networking to study chemistry-based microbial interactions in this system. MALDI IMS was used to visualize the distribution of antimicrobials at the inhibition zone between bacteria associated to the ant Acromyrmex echinatior and the fungal pathogen Escovopsis sp. MS/MS molecular networking was used for the dereplication of compounds found at the inhibition zones. We identified the antibiotics actinomycins D, X2 and X_0β, produced by the bacterium Streptomyces CBR38; and the macrolides elaiophylin, efomycin A and efomycin G, produced by the bacterium Streptomyces CBR53.These metabolites were found at the inhibition zones using MALDI IMS and were identified using MS/MS molecular networking. Additionally, three shearinines D, F, and J produced by the fungal pathogen Escovopsis TZ49 were detected. This is the first report of elaiophylins, actinomycin X_0β and shearinines in the fungus-growing ant symbiotic system. These results suggest a secondary prophylactic use of these antibiotics by A. echinatior because of their permanent production by the bacteria.

Single-molecule protein identification by sub-nanopore sensors

Recent advances in top-down mass spectrometry enabled identification of intact proteins, but this technology still faces challenges. For example, top-down mass spectrometry suffers from a lack of sensitivity since the ion counts for a single fragmentation event are often low. In contrast, nanopore technology is exquisitely sensitive to single intact molecules, but it has only been successfully applied to DNA sequencing, so far. Here, we explore the potential of sub-nanopores for single-molecule protein identification (SMPI) and describe an algorithm for identification of the electrical current blockade signal (nanospectrum) resulting from the translocation of a denaturated, linearly charged protein through a sub-nanopore. The analysis of identification p-values suggests that the current technology is already sufficient for matching nanospectra against small protein databases, e.g., protein identification in bacterial proteomes.

Protein identification is the key step in many proteomics studies. Currently, the most popular technique for intact protein analysis is top-down mass spectrometry which recently enabled high-throughput identification of many proteins and their proteoforms. However, this approach requires large amounts of materials and is currently limited to short proteins, typically less than 30 kDa. On the other hand, nanopore sensors promise single molecule sensitivity in protein analysis, but an approach for the identification of a single protein from its blockade current (nanospectrum) has remained elusive, since the signal from the sensors relates to the amino acid sequence of the protein in a poorly understood way. In this work we describe the first algorithm for protein identification based on nanospectra associated with translocation of proteins through pores with sub-nanometer diameters. While identification accuracy currently does not allow reliable processing of complex protein samples yet, we believe, that the rapidly improving experimental protocols along with the new computational algorithms will transform into a viable protein identification approach in the near future.

New targeted approaches for the quantification of data-independent acquisition mass spectrometry

The use of data-independent acquisition (DIA) approaches for the reproducible and precise quantification of complex protein samples has increased in the last years. The protein information arising from DIA analysis is stored in digital protein maps (DIA maps) that can be interrogated in a targeted way by using ad hoc or publically available peptide spectral libraries generated on the same sample species as for the generation of the DIA maps. The restricted availability of certain difficult-to-obtain human tissues (i.e., brain) together with the caveats of using spectral libraries generated under variable experimental conditions limits the potential of DIA. Therefore, DIA workflows would benefit from high-quality and extended spectral libraries that could be generated without the need of using valuable samples for library production. We describe here two new targeted approaches, using either classical data-dependent acquisition repositories (not specifically built for DIA) or ad hoc mouse spectral libraries, which enable the profiling of human brain DIA data set. The comparison of our results to both the most extended publically available human spectral library and to a state-of-the-art untargeted method supports the use of these new strategies to improve future DIA profiling efforts.

Microbiome Tools for Forensic Science

Microbes are present at every crime scene and have been used as physical evidence for over a century. Advances in DNA sequencing and computational approaches have led to recent breakthroughs in the use of microbiome approaches for forensic science, particularly in the areas of estimating postmortem intervals (PMIs), locating clandestine graves, and obtaining soil and skin trace evidence. Low-cost, high-throughput technologies allow us to accumulate molecular data quickly and to apply sophisticated machine-learning algorithms, building generalizable predictive models that will be useful in the criminal justice system. In particular, integrating microbiome and metabolomic data has excellent potential to advance microbial forensics.

Cardiovascular proteomics in the era of big data: experimental and computational advances

Proteomics plays an increasingly important role in our quest to understand cardiovascular biology. Fueled by analytical and computational advances in the past decade, proteomics applications can now go beyond merely inventorying protein species, and address sophisticated questions on cardiac physiology. The advent of massive mass spectrometry datasets has in turn led to increasing intersection between proteomics and big data science. Here we review new frontiers in technological developments and their applications to cardiovascular medicine. The impact of big data science on cardiovascular proteomics investigations and translation to medicine is highlighted.

Lifestyle chemistries from phones for individual profiling

Imagine a scenario where personal belongings such as pens, keys, phones, or handbags are found at an investigative site. It is often valuable to the investigative team that is trying to trace back the belongings to an individual to understand their personal habits, even when DNA evidence is also available. Here, we develop an approach to translate chemistries recovered from personal objects such as phones into a lifestyle sketch of the owner, using mass spectrometry and informatics approaches. Our results show that phones' chemistries reflect a personalized lifestyle profile. The collective repertoire of molecules found on these objects provides a sketch of the lifestyle of an individual by highlighting the type of hygiene/beauty products the person uses, diet, medical status, and even the location where this person may have been. These findings introduce an additional form of trace evidence from skin-associated lifestyle chemicals found on personal belongings. Such information could help a criminal investigator narrowing down the owner of an object found at a crime scene, such as a suspect or missing person.

Large scale analysis of amino acid substitutions in bacterial proteomics

BACKGROUND

Proteomics of bacterial pathogens is a developing field exploring microbial physiology, gene expression and the complex interactions between bacteria and their hosts. One of the complications in proteomic approach is micro- and macro-heterogeneity of bacterial species, which makes it impossible to build a comprehensive database of bacterial genomes for identification, while most of the existing algorithms rely largely on genomic data.

RESULTS

Here we present a large scale study of identification of single amino acid polymorphisms between bacterial strains. An ad hoc method was developed based on MS/MS spectra comparison without the support of a genomic database. Whole-genome sequencing was used to validate the accuracy of polymorphism detection. Several approaches presented earlier to the proteomics community as useful for polymorphism detection were tested on isolates of Helicobacter pylori, Neisseria gonorrhoeae and Escherichia coli.

CONCLUSION

The developed method represents a perspective approach in the field of bacterial proteomics allowing to identify hundreds of peptides with novel SAPs from a single proteome.

Recognizing millions of consistently unidentified spectra across hundreds of shotgun proteomics datasets

Mass spectrometry (MS) is the main technology used in proteomics approaches. However, on average 75% of spectra analysed in an MS experiment remain unidentified. We propose to use spectrum clustering at a large-scale to shed a light on these unidentified spectra. PRoteomics IDEntifications database (PRIDE) Archive is one of the largest MS proteomics public data repositories worldwide. By clustering all tandem MS spectra publicly available in PRIDE Archive, coming from hundreds of datasets, we were able to consistently characterize three distinct groups of spectra: 1) incorrectly identified spectra, 2) spectra correctly identified but below the set scoring threshold, and 3) truly unidentified spectra. Using a multitude of complementary analysis approaches, we were able to identify less than 20% of the consistently unidentified spectra. The complete spectrum clustering results are available through the new version of the PRIDE Cluster resource (http://www.ebi.ac.uk/pride/cluster). This resource is intended, among other aims, to encourage and simplify further investigation into these unidentified spectra.