Dataset Information

Cross-platform analysis of cancer biomarkers: a Bayesian network approach to incorporating mass spectrometry and microarray data.

ABSTRACT: Many studies showed inconsistent cancer biomarkers due to bioinformatics artifacts. In this paper we use multiple data sets from microarrays, mass spectrometry, protein sequences, and other biological knowledge in order to improve the reliability of cancer biomarkers. We present a novel Bayesian network (BN) model which integrates and cross-annotates multiple data sets related to prostate cancer. The main contribution of this study is that we provide a method that is designed to find cancer biomarkers whose presence is supported by multiple data sources and biological knowledge. Relevant biological knowledge is explicitly encoded into the model parameters, and the biomarker finding problem is formulated as a Bayesian inference problem. Besides diagnostic accuracy, we introduce reliability as another quality measurement of the biological relevance of biomarkers. Based on the proposed BN model, we develop an empirical scoring scheme and a simulation algorithm for inferring biomarkers. Fourteen genes/proteins including prostate specific antigen (PSA) are identified as reliable serum biomarkers which are insensitive to the model assumptions. The computational results show that our method is able to find biologically relevant biomarkers with highest reliability while maintaining competitive predictive power. In addition, by combining biological knowledge and data from multiple platforms, the number of putative biomarkers is greatly reduced to allow more-focused clinical studies.

SUBMITTER: Deng X

PROVIDER: S-EPMC2675834 | biostudies-other | 2007 Apr

REPOSITORIES: biostudies-other

ACCESS DATA

Publications

Cross-platform analysis of cancer biomarkers: a Bayesian network approach to incorporating mass spectrometry and microarray data.

Deng Xutao X Geng Huimin H Ali Hesham H HH

Cancer informatics 20070429

Many studies showed inconsistent cancer biomarkers due to bioinformatics artifacts. In this paper we use multiple data sets from microarrays, mass spectrometry, protein sequences, and other biological knowledge in order to improve the reliability of cancer biomarkers. We present a novel Bayesian network (BN) model which integrates and cross-annotates multiple data sets related to prostate cancer. The main contribution of this study is that we provide a method that is designed to find cancer biom ...[more]

PMID: 19455243

Similar Datasets

Project description:The ability to quantitatively probe diverse panels of proteins and their post-translational modifications (PTMs) across multiple samples would aid a broad spectrum of biological, biochemical and pharmacological studies. We report a novel, microarray analytical technology that combines immuno-affinity capture with Matrix Assisted Laser Desorption Ionization Mass Spectrometry (MALDI MS), which is capable of supporting highly multiplexed, targeted proteomic assays. Termed "Affinity-Bead Assisted Mass Spectrometry" (Affi-BAMS), this LC-free technology enables development of highly specific and customizable assay panels for simultaneous profiling of multiple proteins and PTMs. While affinity beads have been used previously in combination with MS, the Affi-BAMS workflow uses enrichment on a single bead that contains one type of antibody, generally capturing a single analyte (protein or PTM) while having enough binding capacity to enable quantification within approximately 3 orders of magnitude. The multiplexing capability is achieved by combining Affi-BAMS beads with different protein specificities. To enable screening of bead-captured analytes by MS, we further developed a novel method of performing spatially localized elution of targets from individual beads arrayed on a microscope slide. The resulting arrays of micro spots contain highly concentrated analytes localized within 0.5 mm diameter spots that can be directly measured using MALDI MS. While both intact proteins and protein fragments can be monitored by Affi-BAMS, we initially focused on applying this technology for bottom-up proteomics to enable screening of hundreds of samples per day by combining the robust magnetic bead-based workflow with the high throughput nature of MALDI MS acquisition. To demonstrate the variety of applications and robustness of Affi-BAMS, several studies are presented that focus on the response of 4EBP1, RPS6, ERK1/ERK2, mTOR, Histone H3 and C-MET to stimuli including rapamycin, H2O2, EPO, SU11274, Staurosporine and Vorinostat.

Project description:Protein cross-linking and the analysis of cross-linked peptides by mass spectrometry is currently receiving much attention. Not only is this approach applied to isolated complexes to provide information about spatial arrangements of proteins, but it is also increasingly applied to entire cells and their organelles. As in quantitative proteomics, the application of isotopic labeling further makes it possible to monitor quantitative changes in the protein-protein interactions between different states of a system. Here, we cross-linked mitochondria from Saccharomyces cerevisiae grown on either glycerol- or glucose-containing medium to monitor protein-protein interactions under non-fermentative and fermentative conditions. We investigated qualitatively the protein-protein interactions of the 400 most abundant proteins applying stringent data-filtering criteria, i.e. a minimum of two cross-linked peptide spectrum matches and a cut-off in the spectrum scoring of the used search engine. The cross-linker BS3 proved to be equally suited for connecting proteins in all compartments of mitochondria when compared with its water-insoluble but membrane-permeable derivative DSS. We also applied quantitative cross-linking to mitochondria of both the growth conditions using stable-isotope labeled BS3. Significant differences of cross-linked proteins under glycerol and glucose conditions were detected, however, mainly because of the different copy numbers of these proteins in mitochondria under both the conditions. Results obtained from the glycerol condition indicate that the internal NADH:ubiquinone oxidoreductase Ndi1 is part of an electron transport chain supercomplex. We have also detected several hitherto uncharacterized proteins and identified their interaction partners. Among those, Min8 was found to be associated with cytochrome c oxidase. BN-PAGE analyses of min8Δ mitochondria suggest that Min8 promotes the incorporation of Cox12 into cytochrome c oxidase.

Project description:Historically, the database search algorithms have been the de facto standard for inferring peptides from mass spectrometry (MS) data. Database search algorithms deduce peptides by transforming theoretical peptides into theoretical spectra and matching them to the experimental spectra. Heuristic similarity-scoring functions are used to match an experimental spectrum to a theoretical spectrum. However, the heuristic nature of the scoring functions and the simple transformation of the peptides into theoretical spectra, along with noisy mass spectra for the less abundant peptides, can introduce a cascade of inaccuracies. In this paper, we design and implement a Deep Cross-Modal Similarity Network called SpeCollate, which overcomes these inaccuracies by learning the similarity function between experimental spectra and peptides directly from the labeled MS data. SpeCollate transforms spectra and peptides into a shared Euclidean subspace by learning fixed size embeddings for both. Our proposed deep-learning network trains on sextuplets of positive and negative examples coupled with our custom-designed SNAP-loss function. Online hardest negative mining is used to select the appropriate negative examples for optimal training performance. We use 4.8 million sextuplets obtained from the NIST and MassIVE peptide libraries to train the network and demonstrate that for closed search, SpeCollate is able to perform better than Crux and MSFragger in terms of the number of peptide-spectrum matches (PSMs) and unique peptides identified under 1% FDR for real-world data. SpeCollate also identifies a large number of peptides not reported by either Crux or MSFragger. To the best of our knowledge, our proposed SpeCollate is the first deep-learning network that can determine the cross-modal similarity between peptides and mass-spectra for MS-based proteomics. We believe SpeCollate is significant progress towards developing machine-learning solutions for MS-based omics data analysis. SpeCollate is available at https://deepspecs.github.io/.

Dataset Information

Cross-platform analysis of cancer biomarkers: a Bayesian network approach to incorporating mass spectrometry and microarray data.

Publications

Cross-platform analysis of cancer biomarkers: a Bayesian network approach to incorporating mass spectrometry and microarray data.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets