Vasor

'Vasor' is a software tool developed for the assessment of missense variants in the multidrug resistance protein 3 (MDR3), a crucial hepatobiliary transport protein. Dysfunction of MDR3 is associated with various liver diseases, ranging from severe progressive familial intrahepatic cholestasis to transient forms of intrahepatic cholestasis of pregnancy and familial gallstone disease. Many of these diseases can be attributed to single amino acid substitutions in MDR3. However, identifying the effects of these substitutions in in vitro studies can be time-consuming and expensive.
'Vasor' is designed to address this challenge by providing a machine learning-based model for classifying novel missense variants in MDR3 into two categories: benign or pathogenic. It was trained on the largest specific dataset for benign and pathogenic MDR3 variants and utilizes a combination of general predictors, including Evolutionary Models of Variant Effects (EVE), EVmutation, PolyPhen-2, I-Mutant2.0, MUpro, MAESTRO, and PON-P2. It also considers various variant properties, such as half-sphere exposure and posttranslational modification sites.
The Authors state that the tool consistently outperforms integrated general predictors and an external prediction tool called MutPred2, making it the best-performing prediction tool for single-site missense variants in MDR3. 'Vasor' offers comprehensive coverage of the entire sequence space of MDR3, making it a resource for rapidly assessing whether a missense variant is benign or pathogenic. It is accessible as a web server.