Reverse homology is a method for discovering important features in intrinsically disordered regions.
We trained a neural network to identify molecular features of intrinsically disordered regions conserved across evolution. By applying interpretation techniques, we can visualize features that our neural network hypothesizes will be evolutionarily conserved (and thus likely functionally important) features of single sequences.
For more details on how to interpret the outputs of this webtool, check out the Tutorials page.
Questions? Check out our FAQ.
For more details on the method and models, check out our paper in PLOS Computational Biology. If you find this tool useful, we'd appreciate citations!