Amino acid mutations in human visual pigments can impair color vision or lead to diseases such as degenerative blinding condition. Human vision requires that these pigments, consisting of a chromophore and associated opsin protein, have distinct peak spectral sensitivities in separate rod and cone photoreceptor populations. Peak spectral sensitivity is determined by the chromophore type and the amino acid sequence of the opsin. Understanding how a single mutation in the opsin protein can lead to anomalous visual function and disease would assist the development of molecular-level therapeutic strategies. Such an understanding is currently not available. The proposed research has an overarching goal of developing novel molecular-level therapeutic strategies to treat human vision deficiencies by unraveling the mysteries of the phototransduction cycle using state-of-the-art modeling approaches. The goal of the proposed research is to build on our molecular modeling framework to develop and test an automated computational pipeline to estimate peak spectral sensitivity for Rh1 rod opsins and use it to elucidate mechanisms for disease-associated mutations. In Aim 1, we will build a machine-learning-based pipeline, which will utilize homology modeling and molecular dynamics simulation, for accurately predicting peak spectral sensitivity from opsin amino acid sequence data. Aim 2 will employ mixed quantum mechanics / molecular mechanics simulations to elucidate molecular mechanisms of spectral shift and improve the pipeline. In Aim 3, we will use the pipeline to determine molecular mechanisms for anomalous visual functions. The proposed research has the potential for high impact in the field of human vision because it will provide a predictive modeling approach for visual pigments, that has remained elusive for decades. This new genome-to-phenome understanding of the molecular function of visual pigments will pave the way for novel strategies to engineer pigments suitable for optogenetics, or to develop targeted therapeutic strategies.