A multi-model statistical approach for proteomic spectral count quantitation.
ABSTRACT: The rapid development of mass spectrometry (MS) technologies has solidified shotgun proteomics as the most powerful analytical platform for large-scale proteome interrogation. The ability to map and determine differential expression profiles of the entire proteome is the ultimate goal of shotgun proteomics. Label-free quantitation has proven to be a valid approach for discovery shotgun proteomics, especially when sample is limited. Label-free spectral count quantitation is an approach analogous to RNA sequencing whereby count data is used to determine differential expression. Here we show that statistical approaches developed to evaluate differential expression in RNA sequencing experiments can be applied to detect differential protein expression in label-free discovery proteomics. This approach, termed MultiSpec, utilizes open-source statistical platforms; namely edgeR, DESeq and baySeq, to statistically select protein candidates for further investigation. Furthermore, to remove bias associated with a single statistical approach a single ranked list of differentially expressed proteins is assembled by comparing edgeR and DESeq q-values directly with the false discovery rate (FDR) calculated by baySeq. This statistical approach is then extended when applied to spectral count data derived from multiple proteomic pipelines. The individual statistical results from multiple proteomic pipelines are integrated and cross-validated by means of collapsing protein groups.Spectral count data from shotgun proteomics experiments is semi-quantitative and semi-random, yet a robust way to estimate protein concentration. Tag-count approaches are routinely used to analyze RNA sequencing data sets. This approach, termed MultiSpec, utilizes multiple tag-count based statistical tests to determine differential protein expression from spectral counts. The statistical results from these tag-count approaches are combined in order to reach a final MultiSpec q-value to re-rank protein candidates. This re-ranking procedure is completed to remove bias associated with a single approach in order to better understand the true proteomic differences driving the biology in question. The MultiSpec approach can be extended to multiple proteomic pipelines. In such an instance, MultiSpec statistical results are integrated by collapsing protein groups across proteomic pipelines to provide a single ranked list of differentially expressed proteins. This integration mechanism is seamlessly integrated with the statistical analysis and provides the means to cross-validate protein inferences from multiple proteomic pipelines.
Project description:BACKGROUND: Differential expression analysis based on "next-generation" sequencing technologies is a fundamental means of studying RNA expression. We recently developed a multi-step normalization method (called TbT) for two-group RNA-seq data with replicates and demonstrated that the statistical methods available in four R packages (edgeR, DESeq, baySeq, and NBPSeq) together with TbT can produce a well-ranked gene list in which true differentially expressed genes (DEGs) are top-ranked and non-DEGs are bottom ranked. However, the advantages of the current TbT method come at the cost of a huge computation time. Moreover, the R packages did not have normalization methods based on such a multi-step strategy. RESULTS: TCC (an acronym for Tag Count Comparison) is an R package that provides a series of functions for differential expression analysis of tag count data. The package incorporates multi-step normalization methods, whose strategy is to remove potential DEGs before performing the data normalization. The normalization function based on this DEG elimination strategy (DEGES) includes (i) the original TbT method based on DEGES for two-group data with or without replicates, (ii) much faster methods for two-group data with or without replicates, and (iii) methods for multi-group comparison. TCC provides a simple unified interface to perform such analyses with combinations of functions provided by edgeR, DESeq, and baySeq. Additionally, a function for generating simulation data under various conditions and alternative DEGES procedures consisting of functions in the existing packages are provided. Bioinformatics scientists can use TCC to evaluate their methods, and biologists familiar with other R packages can easily learn what is done in TCC. CONCLUSION: DEGES in TCC is essential for accurate normalization of tag count data, especially when up- and down-regulated DEGs in one of the samples are extremely biased in their number. TCC is useful for analyzing tag count data in various scenarios ranging from unbiased to extremely biased differential expression. TCC is available at http://www.iu.a.u-tokyo.ac.jp/~kadota/TCC/ and will appear in Bioconductor (http://bioconductor.org/) from ver. 2.13.
Project description:Differential shotgun proteomics identifies proteins that discriminate between sets of samples based on differences in abundance. This methodology can be easily applied to study (i) specific microorganisms subjected to a variety of growth or stress conditions or (ii) different microorganisms sampled in the same condition. In microbiology, this comparison is particularly successful because differing microorganism phenotypes are explained by clearly altered abundances of key protein players. The extensive description and quantification of proteins from any given microorganism can be routinely obtained for several conditions within a few days by tandem mass spectrometry. Such protein-centred microbial molecular phenotyping is rich in information. However, well-designed experimental strategies, carefully parameterized analytical pipelines, and sound statistical approaches must be applied if the shotgun proteomic data are to be correctly interpreted. This minireview describes these key items for a quick molecular phenotyping based on label-free quantification shotgun proteomics.
Project description:Empirical Bayes is a choice framework for differential expression (DE) analysis for multi-group RNA-seq count data. Its characteristic ability to compute posterior probabilities for predefined expression patterns allows users to assign the pattern with the highest value to the gene under consideration. However, current Bayesian methods such as baySeq and EBSeq can be improved, especially with respect to normalization. Two R packages (baySeq and EBSeq) with their default normalization settings and with other normalization methods (MRN and TCC) were compared using three-group simulation data and real count data. Our findings were as follows: (1) the Bayesian methods coupled with TCC normalization performed comparably or better than those with the default normalization settings under various simulation scenarios, (2) default DE pipelines provided in TCC that implements a generalized linear model framework was still superior to the Bayesian methods with TCC normalization when overall degree of DE was evaluated, and (3) baySeq with TCC was robust against different choices of possible expression patterns. In practice, we recommend using the default DE pipeline provided in TCC for obtaining overall gene ranking and then using the baySeq with TCC normalization for assigning the most plausible expression patterns to individual genes.
Project description:Shotgun proteomics provides the most powerful analytical platform for global inventory of complex proteomes using liquid chromatography-tandem mass spectrometry (LC-MS/MS) and allows a global analysis of protein changes. Nevertheless, sampling of complex proteomes by current shotgun proteomics platforms is incomplete, and this contributes to variability in assessment of peptide and protein inventories by spectral counting approaches. Thus, shotgun proteomics data pose challenges in comparing proteomes from different biological states. We developed an analysis strategy using quasi-likelihood Generalized Linear Modeling (GLM), included in a graphical interface software package (QuasiTel) that reads standard output from protein assemblies created by IDPicker, an HTML-based user interface to query shotgun proteomic data sets. This approach was compared to four other statistical analysis strategies: Student t test, Wilcoxon rank test, Fisher's Exact test, and Poisson-based GLM. We analyzed the performance of these tests to identify differences in protein levels based on spectral counts in a shotgun data set in which equimolar amounts of 48 human proteins were spiked at different levels into whole yeast lysates. Both GLM approaches and the Fisher Exact test performed adequately, each with their unique limitations. We subsequently compared the proteomes of normal tonsil epithelium and HNSCC using this approach and identified 86 proteins with differential spectral counts between normal tonsil epithelium and HNSCC. We selected 18 proteins from this comparison for verification of protein levels between the individual normal and tumor tissues using liquid chromatography-multiple reaction monitoring mass spectrometry (LC-MRM-MS). This analysis confirmed the magnitude and direction of the protein expression differences in all 6 proteins for which reliable data could be obtained. Our analysis demonstrates that shotgun proteomic data sets from different tissue phenotypes are sufficiently rich in quantitative information and that statistically significant differences in proteins spectral counts reflect the underlying biology of the samples.
Project description:RNA-seq, has recently become an attractive method of choice in the studies of transcriptomes, promising several advantages compared with microarrays. In this study, we sought to assess the contribution of the different analytical steps involved in the analysis of RNA-seq data generated with the Illumina platform, and to perform a cross-platform comparison based on the results obtained through Affymetrix microarray. As a case study for our work we, used the Saccharomyces cerevisiae strain CEN.PK 113-7D, grown under two different conditions (batch and chemostat). Here, we asses the influence of genetic variation on the estimation of gene expression level using three different aligners for read-mapping (Gsnap, Stampy and TopHat) on S288c genome, the capabilities of five different statistical methods to detect differential gene expression (baySeq, Cuffdiff, DESeq, edgeR and NOISeq) and we explored the consistency between RNA-seq analysis using reference genome and de novo assembly approach. High reproducibility among biological replicates (correlation?0.99) and high consistency between the two platforms for analysis of gene expression levels (correlation?0.91) are reported. The results from differential gene expression identification derived from the different statistical methods, as well as their integrated analysis results based on gene ontology annotation are in good agreement. Overall, our study provides a useful and comprehensive comparison between the two platforms (RNA-seq and microrrays) for gene expression analysis and addresses the contribution of the different steps involved in the analysis of RNA-seq data.
Project description:Replicate mass spectrometry (MS) measurements and the use of multiple analytical methods can greatly expand the comprehensiveness of shotgun proteomic profiling of biological samples. However, the inherent biases and variations in such data create computational and statistical challenges for quantitative comparative analysis. We developed and tested a normalized, label-free quantitative method termed the normalized spectral index (SI(N)), which combines three MS abundance features: peptide count, spectral count and fragment-ion (tandem MS or MS/MS) intensity. SI(N) largely eliminated variances between replicate MS measurements, permitting quantitative reproducibility and highly significant quantification of thousands of proteins detected in replicate MS measurements of the same and distinct samples. It accurately predicts protein abundance more often than the five other methods we tested. Comparative immunoblotting and densitometry further validate our method. Comparative quantification of complex data sets from multiple shotgun proteomics measurements is relevant for systems biology and biomarker discovery.
Project description:A generally accepted approach to the analysis of RNA-Seq read count data does not yet exist. We sequenced the mRNA of 726 individuals from the Drosophila Genetic Reference Panel in order to quantify differences in gene expression among single flies. One of our experimental goals was to identify the optimal analysis approach for the detection of differential gene expression among the factors we varied in the experiment: genotype, environment, sex, and their interactions. Here we evaluate three different filtering strategies, eight normalization methods, and two statistical approaches using our data set. We assessed differential gene expression among factors and performed a statistical power analysis using the eight biological replicates per genotype, environment, and sex in our data set.We found that the most critical considerations for the analysis of RNA-Seq read count data were the normalization method, underlying data distribution assumption, and numbers of biological replicates, an observation consistent with previous RNA-Seq and microarray analysis comparisons. Some common normalization methods, such as Total Count, Quantile, and RPKM normalization, did not align the data across samples. Furthermore, analyses using the Median, Quantile, and Trimmed Mean of M-values normalization methods were sensitive to the removal of low-expressed genes from the data set. Although it is robust in many types of analysis, the normal data distribution assumption produced results vastly different than the negative binomial distribution. In addition, at least three biological replicates per condition were required in order to have sufficient statistical power to detect expression differences among the three-way interaction of genotype, environment, and sex.The best analysis approach to our data was to normalize the read counts using the DESeq method and apply a generalized linear model assuming a negative binomial distribution using either edgeR or DESeq software. Genes having very low read counts were removed after normalizing the data and fitting it to the negative binomial distribution. We describe the results of this evaluation and include recommended analysis strategies for RNA-Seq read count data.
Project description:BACKGROUND: High-throughput sequencing, such as ribonucleic acid sequencing (RNA-seq) and chromatin immunoprecipitation sequencing (ChIP-seq) analyses, enables various features of organisms to be compared through tag counts. Recent studies have demonstrated that the normalization step for RNA-seq data is critical for a more accurate subsequent analysis of differential gene expression. Development of a more robust normalization method is desirable for identifying the true difference in tag count data. RESULTS: We describe a strategy for normalizing tag count data, focusing on RNA-seq. The key concept is to remove data assigned as potential differentially expressed genes (DEGs) before calculating the normalization factor. Several R packages for identifying DEGs are currently available, and each package uses its own normalization method and gene ranking algorithm. We compared a total of eight package combinations: four R packages (edgeR, DESeq, baySeq, and NBPSeq) with their default normalization settings and with our normalization strategy. Many synthetic datasets under various scenarios were evaluated on the basis of the area under the curve (AUC) as a measure for both sensitivity and specificity. We found that packages using our strategy in the data normalization step overall performed well. This result was also observed for a real experimental dataset. CONCLUSION: Our results showed that the elimination of potential DEGs is essential for more accurate normalization of RNA-seq data. The concept of this normalization strategy can widely be applied to other types of tag count data and to microarray data.
Project description:Analysis pipelines that assign peptides to shotgun proteomics mass spectra often discard identified spectra deemed irrelevant to the scientific hypothesis being tested. To improve statistical power, I propose that researchers remove irrelevant peptides from the database prior to searching rather than assigning these peptides to spectra and then discarding the matches.
Project description:We introduce QPROT, a statistical framework and computational tool for differential protein expression analysis using protein intensity data. QPROT is an extension of the QSPEC suite, originally developed for spectral count data, adapted for the analysis using continuously measured protein-level intensity data. QPROT offers a new intensity normalization procedure and model-based differential expression analysis, both of which account for missing data. Determination of differential expression of each protein is based on the standardized Z-statistic based on the posterior distribution of the log fold change parameter, guided by the false discovery rate estimated by a well-known Empirical Bayes method. We evaluated the classification performance of QPROT using the quantification calibration data from the clinical proteomic technology assessment for cancer (CPTAC) study and a recently published Escherichia coli benchmark dataset, with evaluation of FDR accuracy in the latter.QPROT is a statistical framework with computational software tool for comparative quantitative proteomics analysis. It features various extensions of QSPEC method originally built for spectral count data analysis, including probabilistic treatment of missing values in protein intensity data. With the increasing popularity of label-free quantitative proteomics data, the proposed method and accompanying software suite will be immediately useful for many proteomics laboratories. This article is part of a Special Issue entitled: Computational Proteomics.