S+Proteome

An S-PLUS Module for Protein Mass Spectra Processing and Classification

This project is funded in part with Federal SBIR funds from NCI under contract No. HHSN261200533002C and HHSN261200533003C. The goal is to develop advanced bioinformatics software for analysis and evaluation of proteomic cellular signatures in cancer prevention research. Biomarker discovery for early cancer detection constitutes one of the most important fields in cancer research. Traditional biomarker discovery approaches yield limited sensitivity and specificity in cancer detection. Recent quests to achieve reliable biomarker identification have led researchers to protein analysis (proteomics) via mass spectrometry (MS). Mass spectral signature analysis involves two key steps: processing and classification.

In the first contract, we are addressing the challenges in preprocessing MS data. Recent studies have identified multiple technical and biological shortcomings, resulting in varied forms of noise that must be identified and rectified prior to meaningful use of the data. Insightful's proposed solutions include mass calibration, denoising, baseline detrending, intensity normalization, peak detection, peak alignment, peak quantification, and feature selection and extraction. In the second contract, we are addressing the challenges in classification of high-dimensional MS data to differentiate disease states. Insightful's proposed solutions include a comprehensive suite of cutting-edge classification tools, enabling researchers to explore different techniques, predict classification accuracy, and interpret results in a biologically meaningful way.

The combined software will provide a comprehensive set of tools to assess data quality and state-of-the-art graphical display tools. In addition, we are developing a companion casebook of selected mass spectra data and applications. The software's modular design and extendible structure will facilitate: rigorous validation of promising approaches on extensive datasets; further development of novel bioinformatical techniques; and application of validated biomarker identification disease screening tests.

These projects complement Insightful's S+ArrayAnalyze module for rigorous statistical analysis of microarray data, interactive graphical reporting, automated annotation, publication quality output, and an extensible development environment.

Statistical analysis software