Combining intrinsic disorder prediction and augmented training of hidden Markov models improves discriminative motif discovery

Computational prediction of disordered binding regions

One of the key features of intrinsically disordered regions (IDRs) is their ability to interact with a broad range of partner molecules. Multiple types of interacting IDRs were identified including molecular recognition fragments (MoRFs), short linear sequence motifs (SLiMs), and protein-, nucleic acids- and lipid-binding regions. Prediction of binding IDRs in protein sequences is gaining momentum in recent years. We survey 38 predictors of binding IDRs that target interactions with a diverse set of partners, such as peptides, proteins, RNA, DNA and lipids. We offer a historical perspective and highlight key events that fueled efforts to develop these methods. These tools rely on a diverse range of predictive architectures that include scoring functions, regular expressions, traditional and deep machine learning and meta-models. Recent efforts focus on the development of deep neural network-based architectures and extending coverage to RNA, DNA and lipid-binding IDRs. We analyze availability of these methods and show that providing implementations and webservers results in much higher rates of citations/use. We also make several recommendations to take advantage of modern deep network architectures, develop tools that bundle predictions of multiple and different types of binding IDRs, and work on algorithms that model structures of the resulting complexes.

HH-MOTiF: de novo detection of short linear motifs in proteins by Hidden Markov Model comparisons

Short linear motifs (SLiMs) in proteins are self-sufficient functional sequences that specify interaction sites for other molecules and thus mediate a multitude of functions. Computational, as well as experimental biological research would significantly benefit, if SLiMs in proteins could be correctly predicted de novo with high sensitivity. However, de novo SLiM prediction is a difficult computational task. When considering recall and precision, the performances of published methods indicate remaining challenges in SLiM discovery. We have developed HH-MOTiF, a web-based method for SLiM discovery in sets of mainly unrelated proteins. HH-MOTiF makes use of evolutionary information by creating Hidden Markov Models (HMMs) for each input sequence and its closely related orthologs. HMMs are compared against each other to retrieve short stretches of homology that represent potential SLiMs. These are transformed to hierarchical structures, which we refer to as motif trees, for further processing and evaluation. Our approach allows us to identify degenerate SLiMs, while still maintaining a reasonably high precision. When considering a balanced measure for recall and precision, HH-MOTiF performs better on test data compared to other SLiM discovery methods. HH-MOTiF is freely available as a web-server at http://hh-motif.biochem.mpg.de.