Transmembrane proteins (TMPs) play important roles in cells, ranging from transport processes and cell adhesion to communication. Many of these functions are mediated by intrinsically disordered regions (IDRs), flexible protein segments without a well-defined structure. Although a variety of prediction methods are available for predicting IDRs, their accuracy is highly limited on TMPs due to their special physico-chemical properties.
Methods: We prepared a dataset containing membrane proteins exclusively, using x-ray crystallography data. In addition to attributes commonly used in IDR predictors, we defined several TMP specific features to further enhance the accuracy of our method.
Results: MEMDIS is a novel prediction method, utilizing convolutional neural network and long short-term memory networks for predicting disordered regions in TMPs. MEMDIS achieved the highest prediction accuracy on TMP specific dataset among other popular IDR prediction methods.