Proteogenomic characterization of difficult-to-treat breast cancer with tumor cells enriched through laser microdissection.

BACKGROUND: Breast cancer (BC) is the most commonly diagnosed cancer and the leading cause of cancer death among women globally. Despite advances, there is considerable variation in clinical outcomes for patients with non-luminal A tumors, classified as difficult-to-treat breast cancers (DTBC). This study aims to delineate the proteogenomic landscape of DTBC tumors compared to luminal A (LumA) tumors. METHODS: We retrospectively collected a total of 117 untreated primary breast tumor specimens, focusing on DTBC subtypes. Breast tumors were processed by laser microdissection (LMD) to enrich tumor cells. DNA, RNA, and protein were simultaneously extracted from each tumor preparation, followed by whole genome sequencing, paired-end RNA sequencing, global proteomics and phosphoproteomics. Differential feature analysis, pathway analysis and survival analysis were performed to better understand DTBC and investigate biomarkers. RESULTS: We observed distinct variations in gene mutations, structural variations, and chromosomal alterations between DTBC and LumA breast tumors. DTBC tumors predominantly had more mutations in TP53, PLXNB3, Zinc finger genes, and fewer mutations in SDC2, CDH1, PIK3CA, SVIL, and PTEN. Notably, Cytoband 1q21, which contains numerous cell proliferation-related genes, was significantly amplified in the DTBC tumors. LMD successfully minimized stromal components and increased RNA-protein concordance, as evidenced by stromal score comparisons and proteomic analysis. Distinct DTBC and LumA-enriched clusters were observed by proteomic and phosphoproteomic clustering analysis, some with survival differences. Phosphoproteomics identified two distinct phosphoproteomic profiles for high relapse-risk and low relapse-risk basal-like tumors, involving several genes known to be associated with breast cancer oncogenesis and progression, including KIAA1522, DCK, FOXO3, MYO9B, ARID1A, EPRS, ZC3HAV1, and RBM14. Lastly, an integrated pathway analysis of multi-omics data highlighted a robust enrichment of proliferation pathways in DTBC tumors. CONCLUSIONS: This study provides an integrated proteogenomic characterization of DTBC vs LumA with tumor cells enriched through laser microdissection. We identified many common features of DTBC tumors and the phosphopeptides that could serve as potential biomarkers for high/low relapse-risk basal-like BC and possibly guide treatment selections.

Characterizing Families of Spectral Similarity Scores and Their Use Cases for Gas Chromatography-Mass Spectrometry Small Molecule Identification.

Metabolomics provides a unique snapshot into the world of small molecules and the complex biological processes that govern the human, animal, plant, and environmental ecosystems encapsulated by the One Health modeling framework. However, this "molecular snapshot" is only as informative as the number of metabolites confidently identified within it. The spectral similarity (SS) score is traditionally used to identify compound(s) in mass spectrometry approaches to metabolomics, where spectra are matched to reference libraries of candidate spectra. Unfortunately, there is little consensus on which of the dozens of available SS metrics should be used. This lack of standard SS score creates analytic uncertainty and potentially leads to issues in reproducibility, especially as these data are integrated across other domains. In this work, we use metabolomic spectral similarity as a case study to showcase the challenges in consistency within just one piece of the One Health framework that must be addressed to enable data science approaches for One Health problems. Here, using a large cohort of datasets comprising both standard and complex datasets with expert-verified truth annotations, we evaluated the effectiveness of 66 similarity metrics to delineate between correct matches (true positives) and incorrect matches (true negatives). We additionally characterize the families of these metrics to make informed recommendations for their use. Our results indicate that specific families of metrics (the Inner Product, Correlative, and Intersection families of scores) tend to perform better than others, with no single similarity metric performing optimally for all queried spectra. This work and its findings provide an empirically-based resource for researchers to use in their selection of similarity metrics for GC-MS identification, increasing scientific reproducibility through taking steps towards standardizing identification workflows.

A genomic data archive from the Network for Pancreatic Organ donors with Diabetes.

The Network for Pancreatic Organ donors with Diabetes (nPOD) is the largest biorepository of human pancreata and associated immune organs from donors with type 1 diabetes (T1D), maturity-onset diabetes of the young (MODY), cystic fibrosis-related diabetes (CFRD), type 2 diabetes (T2D), gestational diabetes, islet autoantibody positivity (AAb+), and without diabetes. nPOD recovers, processes, analyzes, and distributes high-quality biospecimens, collected using optimized standard operating procedures, and associated de-identified data/metadata to researchers around the world. Herein describes the release of high-parameter genotyping data from this collection. 372 donors were genotyped using a custom precision medicine single nucleotide polymorphism (SNP) microarray. Data were technically validated using published algorithms to evaluate donor relatedness, ancestry, imputed HLA, and T1D genetic risk score. Additionally, 207 donors were assessed for rare known and novel coding region variants via whole exome sequencing (WES). These data are publicly-available to enable genotype-specific sample requests and the study of novel genotype:phenotype associations, aiding in the mission of nPOD to enhance understanding of diabetes pathogenesis to promote the development of novel therapies.

Better understanding and prediction of antiviral peptides through primary and secondary structure feature importance.

The emergence of viral epidemics throughout the world is of concern due to the scarcity of available effective antiviral therapeutics. The discovery of new antiviral therapies is imperative to address this challenge, and antiviral peptides (AVPs) represent a valuable resource for the development of novel therapies to combat viral infection. We present a new machine learning model to distinguish AVPs from non-AVPs using the most informative features derived from the physicochemical and structural properties of their amino acid sequences. To focus on those features that are most likely to contribute to antiviral performance, we filter potential features based on their importance for classification. These feature selection analyses suggest that secondary structure is the most important peptide sequence feature for predicting AVPs. Our Feature-Informed Reduced Machine Learning for Antiviral Peptide Prediction (FIRM-AVP) approach achieves a higher accuracy than either the model with all features or current state-of-the-art single classifiers. Understanding the features that are associated with AVP activity is a core need to identify and design new AVPs in novel systems. The FIRM-AVP code and standalone software package are available at https://github.com/pmartR/FIRM-AVP with an accompanying web application at https://msc-viz.emsl.pnnl.gov/AVPR .

The Cancer Microbiome: Distinguishing Direct and Indirect Effects Requires a Systemic View.

The collection of microbes that live in and on the human body - the human microbiome - can impact on cancer initiation, progression, and response to therapy, including cancer immunotherapy. The mechanisms by which microbiomes impact on cancers can yield new diagnostics and treatments, but much remains unknown. The interactions between microbes, diet, host factors, drugs, and cell-cell interactions within the cancer itself likely involve intricate feedbacks, and no single component can explain all the behavior of the system. Understanding the role of host-associated microbial communities in cancer systems will require a multidisciplinary approach combining microbial ecology, immunology, cancer cell biology, and computational biology - a systems biology approach.

Quality Control Analysis in Real-time (QC-ART): A Tool for Real-time Quality Control Assessment of Mass Spectrometry-based Proteomics Data.

Liquid chromatography-mass spectrometry (LC-MS)-based proteomics studies of large sample cohorts can easily require from months to years to complete. Acquiring consistent, high-quality data in such large-scale studies is challenging because of normal variations in instrumentation performance over time, as well as artifacts introduced by the samples themselves, such as those because of collection, storage and processing. Existing quality control methods for proteomics data primarily focus on post-hoc analysis to remove low-quality data that would degrade downstream statistics; they are not designed to evaluate the data in near real-time, which would allow for interventions as soon as deviations in data quality are detected. In addition to flagging analyses that demonstrate outlier behavior, evaluating how the data structure changes over time can aide in understanding typical instrument performance or identify issues such as a degradation in data quality because of the need for instrument cleaning and/or re-calibration. To address this gap for proteomics, we developed Quality Control Analysis in Real-Time (QC-ART), a tool for evaluating data as they are acquired to dynamically flag potential issues with instrument performance or sample quality. QC-ART has similar accuracy as standard post-hoc analysis methods with the additional benefit of real-time analysis. We demonstrate the utility and performance of QC-ART in identifying deviations in data quality because of both instrument and sample issues in near real-time for LC-MS-based plasma proteomics analyses of a sample subset of The Environmental Determinants of Diabetes in the Young cohort. We also present a case where QC-ART facilitated the identification of oxidative modifications, which are often underappreciated in proteomic experiments.

High-Fat Diets Alter the Modulatory Effects of Xenobiotics on Cytochrome P450 Activities.

Cytochrome P450 monooxygenase (P450) enzymes metabolize critical endogenous chemicals and oxidize nearly all xenobiotics. Dysregulated P450 activities lead to altered capacity for drug metabolism and cellular stress. The effects of mixed exposures on P450 expression and activity are variable and elusive. A high-fat diet (HFD) is a common exposure that results in obesity and associated pathologies including hepatotoxicity. Herein, we report the effects of cigarette smoke on P450 activities of normal weight and HFD induced obese mice. Activity-based protein profiling results indicate that HFD mice had significantly decreased P450 activity, likely instigated by proinflammatory chemicals, and that P450 enzymes involved in detoxification, xenobiotic metabolism, and bile acid synthesis were effected by HFD and smoke interaction. Smoking increased activity of all lung P450 and coexposure to diet effected P450 2s1. We need to expand our understanding of common exposures coupled to altered P450 metabolism to enhance the safety and efficacy of therapeutic drug dosing.

P-MartCancer-Interactive Online Software to Enable Analysis of Shotgun Cancer Proteomic Datasets.

P-MartCancer is an interactive web-based software environment that enables statistical analyses of peptide or protein data, quantitated from mass spectrometry-based global proteomics experiments, without requiring in-depth knowledge of statistical programming. P-MartCancer offers a series of statistical modules associated with quality assessment, peptide and protein statistics, protein quantification, and exploratory data analyses driven by the user via customized workflows and interactive visualization. Currently, P-MartCancer offers access and the capability to analyze multiple cancer proteomic datasets generated through the Clinical Proteomics Tumor Analysis Consortium at the peptide, gene, and protein levels. P-MartCancer is deployed as a web service (https://pmart.labworks.org/cptac.html), alternatively available via Docker Hub (https://hub.docker.com/r/pnnl/pmart-web/). Cancer Res; 77(21); e47-50. ©2017 AACR.

MPLEx: a method for simultaneous pathogen inactivation and extraction of samples for multi-omics profiling.

The continued emergence and spread of infectious agents is of great concern, and systems biology approaches to infectious disease research can advance our understanding of host-pathogen relationships and facilitate the development of new therapies and vaccines. Molecular characterization of infectious samples outside of appropriate biosafety containment can take place only subsequent to pathogen inactivation. Herein, we describe a modified Folch extraction using chloroform/methanol that facilitates the molecular characterization of infectious samples by enabling simultaneous pathogen inactivation and extraction of proteins, metabolites, and lipids for subsequent mass spectrometry-based multi-omics measurements. This single-sample metabolite, protein and lipid extraction (MPLEx) method resulted in complete inactivation of clinically important bacterial and viral pathogens with exposed lipid membranes, including Yersinia pestis, Salmonella Typhimurium, and Campylobacter jejuni in pure culture, and Yersinia pestis, Campylobacter jejuni, and West Nile, MERS-CoV, Ebola, and influenza H7N9 viruses in infection studies. In addition, >99% inactivation, which increased with solvent exposure time, was also observed for pathogens without exposed lipid membranes including community-associated methicillin-resistant Staphylococcus aureus, Clostridium difficile spores and vegetative cells, and adenovirus type 5. The overall pipeline of inactivation and subsequent proteomic, metabolomic, and lipidomic analyses was evaluated using a human epithelial lung cell line infected with wild-type and mutant influenza H7N9 viruses, thereby demonstrating that MPLEx yields biomaterial of sufficient quality for subsequent multi-omics analyses. Based on these experimental results, we believe that MPLEx will facilitate systems biology studies of infectious samples by enabling simultaneous pathogen inactivation and multi-omics measurements from a single specimen with high success for pathogens with exposed lipid membranes.

Silymarin Suppresses Cellular Inflammation By Inducing Reparative Stress Signaling.

Silymarin, a characterized extract of the seeds of milk thistle (Silybum marianum), suppresses cellular inflammation. To define how this occurs, transcriptional profiling, metabolomics, and signaling studies were performed in human liver and T cell lines. Cellular stress and metabolic pathways were modulated within 4 h of silymarin treatment: activation of Activating Transcription Factor 4 (ATF-4) and adenosine monophosphate protein kinase (AMPK) and inhibition of mammalian target of rapamycin (mTOR) signaling, the latter being associated with induction of DNA-damage-inducible transcript 4 (DDIT4). Metabolomics analyses revealed silymarin suppression of glycolytic, tricarboxylic acid (TCA) cycle, and amino acid metabolism. Anti-inflammatory effects arose with prolonged (i.e., 24 h) silymarin exposure, with suppression of multiple pro-inflammatory mRNAs and signaling pathways including nuclear factor kappa B (NF-κB) and forkhead box O (FOXO). Studies with murine knock out cells revealed that silymarin inhibition of both mTOR and NF-κB was partially AMPK dependent, whereas silymarin inhibition of mTOR required DDIT4. Other natural products induced similar stress responses, which correlated with their ability to suppress inflammation. Thus, natural products activate stress and repair responses that culminate in an anti-inflammatory cellular phenotype. Natural products like silymarin may be useful as tools to define how metabolic, stress, and repair pathways regulate cellular inflammation.

Review, evaluation, and discussion of the challenges of missing value imputation for mass spectrometry-based label-free global proteomics.

In this review, we apply selected imputation strategies to label-free liquid chromatography-mass spectrometry (LC-MS) proteomics datasets to evaluate the accuracy with respect to metrics of variance and classification. We evaluate several commonly used imputation approaches for individual merits and discuss the caveats of each approach with respect to the example LC-MS proteomics data. In general, local similarity-based approaches, such as the regularized expectation maximization and least-squares adaptive algorithms, yield the best overall performances with respect to metrics of accuracy and robustness. However, no single algorithm consistently outperforms the remaining approaches, and in some cases, performing classification without imputation sometimes yielded the most accurate classification. Thus, because of the complex mechanisms of missing data in proteomics, which also vary from peptide to protein, no individual method is a single solution for imputation. On the basis of the observations in this review, the goal for imputation in the field of computational proteomics should be to develop new approaches that work generically for this data type and new strategies to guide users in the selection of the best imputation for their dataset and analysis objectives.

A semiautomated framework for integrating expert knowledge into disease marker identification.

BACKGROUND: The availability of large complex data sets generated by high throughput technologies has enabled the recent proliferation of disease biomarker studies. However, a recurring problem in deriving biological information from large data sets is how to best incorporate expert knowledge into the biomarker selection process. OBJECTIVE: To develop a generalizable framework that can incorporate expert knowledge into data-driven processes in a semiautomated way while providing a metric for optimization in a biomarker selection scheme. METHODS: The framework was implemented as a pipeline consisting of five components for the identification of signatures from integrated clustering (ISIC). Expert knowledge was integrated into the biomarker identification process using the combination of two distinct approaches; a distance-based clustering approach and an expert knowledge-driven functional selection. RESULTS: The utility of the developed framework ISIC was demonstrated on proteomics data from a study of chronic obstructive pulmonary disease (COPD). Biomarker candidates were identified in a mouse model using ISIC and validated in a study of a human cohort. CONCLUSIONS: Expert knowledge can be introduced into a biomarker discovery process in different ways to enhance the robustness of selected marker candidates. Developing strategies for extracting orthogonal and robust features from large data sets increases the chances of success in biomarker identification.

Impaired transcriptional response of the murine heart to cigarette smoke in the setting of high fat diet and obesity.

Smoking and obesity are each well-established risk factors for cardiovascular heart disease, which together impose earlier onset and greater severity of disease. To identify early signaling events in the response of the heart to cigarette smoke exposure within the setting of obesity, we exposed normal weight and high fat diet-induced obese (DIO) C57BL/6 mice to repeated inhaled doses of mainstream (MS) or sidestream (SS) cigarette smoke administered over a two week period, monitoring effects on both cardiac and pulmonary transcriptomes. MS smoke (250 µg wet total particulate matter (WTPM)/L, 5 h/day) exposures elicited robust cellular and molecular inflammatory responses in the lung with 1466 differentially expressed pulmonary genes (p < 0.01) in normal weight animals and a much-attenuated response (463 genes) in the hearts of the same animals. In contrast, exposures to SS smoke (85 µg WTPM/L) with a CO concentration equivalent to that of MS smoke (~250 CO ppm) induced a weak pulmonary response (328 genes) but an extensive cardiac response (1590 genes). SS smoke and to a lesser extent MS smoke preferentially elicited hypoxia- and stress-responsive genes as well as genes predicting early changes of vascular smooth muscle and endothelium, precursors of cardiovascular disease. The most sensitive smoke-induced cardiac transcriptional changes of normal weight mice were largely absent in DIO mice after smoke exposure, while genes involved in fatty acid utilization were unaffected. At the same time, smoke exposure suppressed multiple proteome maintenance genes induced in the hearts of DIO mice. Together, these results underscore the sensitivity of the heart to SS smoke and reveal adaptive responses in healthy individuals that are absent in the setting of high fat diet and obesity.

Release of severe acute respiratory syndrome coronavirus nuclear import block enhances host transcription in human lung cells.

The severe acute respiratory syndrome coronavirus accessory protein ORF6 antagonizes interferon signaling by blocking karyopherin-mediated nuclear import processes. Viral nuclear import antagonists, expressed by several highly pathogenic RNA viruses, likely mediate pleiotropic effects on host gene expression, presumably interfering with transcription factors, cytokines, hormones, and/or signaling cascades that occur in response to infection. By bioinformatic and systems biology approaches, we evaluated the impact of nuclear import antagonism on host expression networks by using human lung epithelial cells infected with either wild-type virus or a mutant that does not express ORF6 protein. Microarray analysis revealed significant changes in differential gene expression, with approximately twice as many upregulated genes in the mutant virus samples by 48 h postinfection, despite identical viral titers. Our data demonstrated that ORF6 protein expression attenuates the activity of numerous karyopherin-dependent host transcription factors (VDR, CREB1, SMAD4, p53, EpasI, and Oct3/4) that are critical for establishing antiviral responses and regulating key host responses during virus infection. Results were confirmed by proteomic and chromatin immunoprecipitation assay analyses and in parallel microarray studies using infected primary human airway epithelial cell cultures. The data strongly support the hypothesis that viral antagonists of nuclear import actively manipulate host responses in specific hierarchical patterns, contributing to the viral pathogenic potential in vivo. Importantly, these studies and modeling approaches not only provide templates for evaluating virus antagonism of nuclear import processes but also can reveal candidate cellular genes and pathways that may significantly influence disease outcomes following severe acute respiratory syndrome coronavirus infection in vivo.

Diet-induced obesity reprograms the inflammatory response of the murine lung to inhaled endotoxin.

The co-occurrence of environmental factors is common in complex human diseases and, as such, understanding the molecular responses involved is essential to determine risk and susceptibility to disease. We have investigated the key biological pathways that define susceptibility for pulmonary infection during obesity in diet-induced obese (DIO) and regular weight (RW) C57BL/6 mice exposed to inhaled lipopolysaccharide (LPS). LPS induced a strong inflammatory response in all mice as indicated by elevated cell counts of macrophages and neutrophils and levels of proinflammatory cytokines (MDC, MIP-1γ, IL-12, RANTES) in the bronchoalveolar lavage fluid. Additionally, DIO mice exhibited 50% greater macrophage cell counts, but decreased levels of the cytokines, IL-6, TARC, TNF-α, and VEGF relative to RW mice. Microarray analysis of lung tissue showed over half of the LPS-induced expression in DIO mice consisted of genes unique for obese mice, suggesting that obesity reprograms how the lung responds to subsequent insult. In particular, we found that obese animals exposed to LPS have gene signatures showing increased inflammatory and oxidative stress response and decreased antioxidant capacity compared with RW. Because signaling pathways for these responses can be common to various sources of environmentally induced lung damage, we further identified biomarkers that are indicative of specific toxicant exposure by comparing gene signatures after LPS exposure to those from a parallel study with cigarette smoke. These data show obesity may increase sensitivity to further insult and that co-occurrence of environmental stressors result in complex biosignatures that are not predicted from analysis of individual exposures.

The effects of low-dose irradiation on inflammatory response proteins in a 3D reconstituted human skin tissue model.

Skin responses to moderate and high doses of ionizing radiation include the induction of DNA repair, apoptosis and stress response pathways. Additionally, numerous studies indicate that radiation exposure leads to inflammatory responses in skin cells and tissue. However, the inflammatory response of skin tissue to low-dose radiation (≤10 cGy) is poorly understood. To address this, we have utilized a reconstituted human skin tissue model (MatTek EpiDermFT™) and assessed changes in 23 cytokines, 24 and 48 h after treatment of skin with either 3 or 10 cGy low dose of radiation. Three cytokines, IFN-γ, IL-2, MIP-1α, were significantly altered in response to low-dose radiation. In contrast, seven cytokines were significantly altered in response to a high radiation dose of 200 cGy (IL-2, IL-10, IL-13, IFN-γ, MIP-1α, TNFα and VEGF) or the tumor promoter 12-O-tetradecanoylphorbol 13-acetate (G-CSF, GM-CSF, IL-1α, IL-8, MIP-1α, MIP-1ß and RANTES). Additionally, radiation induced inflammation appears to have a distinct cytokine response relative to the nonradiation induced stressor, TPA. Overall, these results indicate that there are subtle changes in the inflammatory protein levels after exposure to low-dose radiation and this response is a subset of what is seen after a high dose in a human skin tissue model.

Plasma biomarkers for detecting Hodgkin's lymphoma in HIV patients.

The lifespan of people with human immunodeficiency virus (HIV) infection has increased as a result of effective antiretroviral therapy, and the incidences of the AIDS-defining cancers, non-Hodgkin's lymphoma and Kaposi sarcoma, have declined. Even so, HIV-infected individuals are now at greater risk of other cancers, including Hodgkin's lymphoma (HL). To identify candidate biomarkers for the early detection of HL, we undertook an accurate mass and elution time tag proteomics analysis of individual plasma samples from either HIV-infected patients without HL (controls; n = 14) and from HIV-infected patient samples with HL (n = 22). This analysis identified 60 proteins that were statistically (p<0.05) altered and at least 1.5-fold different between the two groups. At least three of these proteins have previously been reported to be altered in the blood of HL patients that were not known to be HIV positive, suggesting that these markers may be broadly useful for detecting HL. Ingenuity Pathway Analysis software identified "inflammatory response" and "cancer" as the top two biological functions associated with these proteins. Overall, this study validated three plasma proteins as candidate biomarkers for detecting HL, and identified 57 novel candidate biomarkers that remain to be validated. The relationship of these novel candidate biomarkers with cancer and inflammation suggests that they are truly associated with HL and therefore may be useful for the early detection of this cancer in susceptible populations.

A statistical selection strategy for normalization procedures in LC-MS proteomics experiments through dataset-dependent ranking of normalization scaling factors.

Quantification of LC-MS peak intensities assigned during peptide identification in a typical comparative proteomics experiment will deviate from run-to-run of the instrument due to both technical and biological variation. Thus, normalization of peak intensities across an LC-MS proteomics dataset is a fundamental step in pre-processing. However, the downstream analysis of LC-MS proteomics data can be dramatically affected by the normalization method selected. Current normalization procedures for LC-MS proteomics data are presented in the context of normalization values derived from subsets of the full collection of identified peptides. The distribution of these normalization values is unknown a priori. If they are not independent from the biological factors associated with the experiment the normalization process can introduce bias into the data, possibly affecting downstream statistical biomarker discovery. We present a novel approach to evaluate normalization strategies, which includes the peptide selection component associated with the derivation of normalization values. Our approach evaluates the effect of normalization on the between-group variance structure in order to identify the most appropriate normalization methods that improve the structure of the data without introducing bias into the normalized peak intensities.

Improved quality control processing of peptide-centric LC-MS proteomics data.

MOTIVATION: In the analysis of differential peptide peak intensities (i.e. abundance measures), LC-MS analyses with poor quality peptide abundance data can bias downstream statistical analyses and hence the biological interpretation for an otherwise high-quality dataset. Although considerable effort has been placed on assuring the quality of the peptide identification with respect to spectral processing, to date quality assessment of the subsequent peptide abundance data matrix has been limited to a subjective visual inspection of run-by-run correlation or individual peptide components. Identifying statistical outliers is a critical step in the processing of proteomics data as many of the downstream statistical analyses [e.g. analysis of variance (ANOVA)] rely upon accurate estimates of sample variance, and their results are influenced by extreme values. RESULTS: We describe a novel multivariate statistical strategy for the identification of LC-MS runs with extreme peptide abundance distributions. Comparison with current method (run-by-run correlation) demonstrates a significantly better rate of identification of outlier runs by the multivariate strategy. Simulation studies also suggest that this strategy significantly outperforms correlation alone in the identification of statistically extreme liquid chromatography-mass spectrometry (LC-MS) runs. AVAILABILITY: https://www.biopilot.org/docs/Software/RMD.php CONTACT: bj@pnl.gov SUPPLEMENTARY INFORMATION: Supplementary material is available at Bioinformatics online.

Combined statistical analyses of peptide intensities and peptide occurrences improves identification of significant peptides from MS-based proteomics data.

Liquid chromatography-mass spectrometry-based (LC-MS) proteomics uses peak intensities of proteolytic peptides to infer the differential abundance of peptides/proteins. However, substantial run-to-run variability in intensities and observations (presence/absence) of peptides makes data analysis quite challenging. The missing observations in LC-MS proteomics data are difficult to address with traditional imputation-based approaches because the mechanisms by which data are missing are unknown a priori. Data can be missing due to random mechanisms such as experimental error or nonrandom mechanisms such as a true biological effect. We present a statistical approach that uses a test of independence known as a G-test to test the null hypothesis of independence between the number of missing values across experimental groups. We pair the G-test results, evaluating independence of missing data (IMD) with an analysis of variance (ANOVA) that uses only means and variances computed from the observed data. Each peptide is therefore represented by two statistical confidence metrics, one for qualitative differential observation and one for quantitative differential intensity. We use three LC-MS data sets to demonstrate the robustness and sensitivity of the IMD-ANOVA approach.