The comprehensive and quantitative analysis of clinical proteomic samples is an outstanding challenge in biomedical research. New proteomic technologies for cancer detection are urgently needed and hold great potential for improving human health, as underscored by the improved survival rates of patients diagnosed in he early stages of cancer. To this end, we will develop computational tools aimed at increasing the effectiveness of cancer biomarker discovery from label-free MALDI-TOF (matrix-assisted laser- desorption/ionization time-of-flight) mass spectra for verification and identification. The computational algorithms and tools will result in more than an order of magnitude increase in both sensitivity and selectivity For molecular biomarker screening. Specifically, we propose: (i) to optimize signal processing resulting in at east a 4-fold enhancement of sensitivity (as measured by signal-to-noise), 2-fold gain in selectivity (resolution), and 10-fold increase in mass accuracy (Aim 1);(ii) to automate detection of ionization satellite ons followed by mass recalibration (Aim 2) resulting in tripling selectivity and mass accuracy;(iii) to deconvolve intensity distributions from satellite ions into parent protein peaks (Aim 3) resulting in tripling sensitivity for statistical detection and experimental identification of biomarkers from enhanced molecular maps (Aim 4). The increased efficiency of broad mass range screening will decrease the time and cost of the downstream identification and validation experiments. The successful completion of the studies described in this application will provide a basis for expanding these computational tools to other TOP MS platforms, and advance the endeavor of characterizing molecular basis for cancer toward better prognosis and treatment strategies.
