Background: A number of observational and intervention studies have investigated the relationship between physical activity and mental health; however, few studies evaluate the association between physical activity and depression in a population sample by using minute by minute data over one week from accelerometer. The purpose of this study is to explore the different physical activity patterns and the relationships between these patterns and depression symptoms based on minute by minute accelerometer assessed data. Methods: Data from the Mitchelstown cohort study were used. Taking consider of non-wear time and background information missing, 375 participants were included in this study. They all completed questionnaires and wore accelerometers for seven consecutive days. Questionnaire provided background information and Center for Epidemiological Studies Depression Scale (CES-D) score measurements, accelerometer output provided minute by minute physical activity data. Bivariate smoothing method was used to explore the interaction effect of depression score and other continuous background covariates to CESD, multiple regression analysis was used to get the relationship between CES-D score and physical activity level. Results: Within Day Physical activity profile analysis showed that after 11: 00 pm and before around 7: 00 am, participants in moderate and moderate to severe depression groups are much active than the other two groups, but during the other day time, moderate to severe group is less active than the others. There were strong contrasts between depression groups regarding time-of day of peak per minute activity. Daily activity gets progressively lower for moderate to severe group since between 7 am and 8 am, and the cumulative activity is the lowest among these four groups. Bivariate relationship analysis also showed there were difference between male and female, different depression groups participants.