Traditionally, early detection of prostate cancer occurs by increased prostate-specific antigen (PSA) in blood and then prostatic carcinoma (PCa) is confirmed using histopathological biopsy samples under a microscope.But PSA is prostate specific rather than prostate cancer specific so increased PSA occurs in other conditions such as benign prostatic hyperplasia--resulting in many unnecessary biopsies.Therefore, new proteomic biomarkers are needed.However, analysis of protein expressions and interactions to pick biomarkers is very challenging due to its complexity and noise from unrelated proteins.To improve analyses, we identify textural features (optical biomarkers) to classify high resolution histopathological images based on pathologist-guided low resolution microscope imagery.Then,we utilize this information from the optical images to map the proteomic analysis of MALDI (Matrix-Assisted Laser Desorption/Ionization) mass spectrometry data, which identify a protein or protein combination (molecular biomarkers) that best classifies prostate cancer vs normal regions.Our hybrid molecular and optical imaging approach is constrained by the H&E fixation of the fresh frozen biopsies that render the tissue unusable for MALD1 testing.In our three step method, we first apply a texture analysis technique on the high magnification optical image to predict PCa regions on an adjacent tissue slice.Separately, we analyze the MALDI data from the accompanying tissue slice.Finally, we combine both results to obtain a PCa region that can improve biomarker identification.Results show that the texture-based prediction is sensitive but less specific and the MALDI-based prediction is specific but not sensitive.When combined, an improved prediction for PCa regions on the accompanying slice can be achieved.We believe this prediction allows a better informed search for enhanced PCa biomarkers.