Transgender individuals experience discrimination in all domains of their personal and social life. Discrimination is believed to be associated with worse quality of life (QoL). To investigate the relation between QoL and perceived levels of discrimination and social support in individuals with gender dysphoria (GD). Individuals with GD who attended a psychiatry clinic from January 2012 through December 2014 were recruited. Demographic, social, and medical transition features were collected with standardized forms. Self-report measurements of QoL (Turkish version of the World Health Organization's Quality of Life-BREF) that included physical, psychological, social, and environmental domains, perceived discrimination with personal and group subscales (Perceived Discrimination Scale [PDS]), and social support (Multidimensional Scale of Perceived Social Support) were completed. Ninety-four participants (76.6% trans men) adequately completed the study measurements. Regression models with each QoL domain score as a dependent variable indicated a significant predictor value of personal PDS in social and environmental QoL. Social support from family was associated with better QoL in psychological QoL, whereas perceived support from friends significantly predicted all other domains of QoL. There was a tendency for group PDS to be rated higher than personal PDS, suggesting personal vs group discrimination discrepancy. However, group PDS was not found to be a predictor of QoL in the multivariate model. Perceived personal discrimination and social support from different sources predicted domains of QoL with a non-uniform pattern in individuals with GD. Social support and discrimination were found to have opposing contributions to QoL in GD. The present findings emphasize the necessity of addressing discrimination and social support in clinical work with GD. Moreover, strategies to improve and strengthen friend and family support for individuals with GD should be explored by clinicians. Further research with larger and community-based samples is required.