We present a computational analysis of Mass Spectrometry (MS) data based on a proteomic study of mice cardiac tissue samples whose measured changes in peptide and protein abundance were obtained under normal (control or CTRL) and simulated microgravity conditions (hind-limb unloading or HLU). A pipeline consisting of experimental and computational steps has been designed to achieve, respectively, pre-fractionation to simplify mouse heart protein extracts and data synthesis by feature consensus maps. Both acid and neutral protein fractions obtained from differential solubility have been digested, and peptide mixtures have been resolved by LC-MALDI. The computational tools employed to analyze the MS data and reduce their complex dimensionality have included spectra alignment, denoising and normalization to obtain good-quality peak detection performance. In turn, features could be identified and further refined by searching patterns across repeated measurements obtained under differential conditions (HLU-CTRL, acid-neutral protein extracts). The assessment of reproducibility aspects for evaluating the efficacy of label-free comparative proteomic analysis, combined with the goal of identifying from HLU-CTRL consensus maps candidate proteins with differential expression, led to a computationally efficient approach delivering proteins related to the basic heart functions, cardiac stress and energy supply.