Abstract: Combination of GIS technology with geological, geophysical, geochemical data is currently an important prospecting trend in comprehensive metallogenic prediction. Dongxi gold ore deposit and peripheral areas at the 1:10 000 scale were selected as the study objective. Based on field investigation, spatial statistic results and mineralization features, a prediction prospecting model was established through extracting geological data, such as geological, geophysical, geochemical information, from GIS platform. Qualified independent prediction factors were selected as evidence layers to analyze the spatial distribution of each layer and ore bodies at a grid cell of 5 m×5 m. We conducted mineralization prediction for the mining area and peripheral areas using weights-of-evidence method and made a comprehensive review over the defined target areas based on the spatial distribution of high-probability mineralization using graphic expression. The results show that the target areas carry promising geological, geophysical and geochemical prospecting data, indicating good gold mineralization information in each target area.