Search

Intrinsically disordered proteins, defying the traditional protein structure–function paradigm, are a challenge to study experimentally. Because a large part of our knowledge rests on computational predictions, it is crucial that their accuracy is high. The Critical Assessment of protein Intrinsic Disorder prediction (CAID) experiment was established as a community-based blind test to determine the state of the art in prediction of intrinsically disordered regions and the subset of residues involved in binding. A total of 43 methods were evaluated on a dataset of 646 proteins from DisProt. The best methods use deep learning techniques and notably outperform physicochemical methods. The top disorder predictor has Fmax = 0.483 on the full dataset and Fmax = 0.792 following filtering out of bona fide structured regions. Disordered binding regions remain hard to predict, with Fmax = 0.231. Interestingly, computing times among methods can vary by up to four orders of magnitude., Results are presented from the first Critical Assessment of protein Intrinsic Disorder prediction (CAID) experiment, a community-based blind test to determine the state of the art in predicting intrinsically disordered regions in proteins.

Short Linear Motifs (SLiMs) are functional protein microdomains that typically mediate interactions between a short linear region in one protein and a globular domain in another. SLiMs usually occur in structurally disordered regions and mediate low affinity interactions. Most SLiMs are 3-15 amino acids in length and have 2-5 defined positions, making them highly likely to occur by chance and extremely difficult to identify. Nevertheless, our knowledge of SLiMs and capacity to predict them from protein sequence data using computational methods has advanced dramatically over the past decade. By considering the biological, structural, and evolutionary context of SLiM occurrences, it is possible to differentiate functional instances from chance matches in many cases and to identify new regions of proteins that have the features consistent with a SLiM-mediated interaction. Their simplicity also makes SLiMs evolutionarily labile and prone to independent origins on different sequence backgrounds through convergent evolution, which can be exploited for predicting novel SLiMs in proteins that share a function or interaction partner. In this review, we explore our current knowledge of SLiMs and how it can be applied to the task of predicting them computationally from protein sequences. Rather than focusing on specific SLiM prediction tools, we provide an overview of the methods available and concentrate on principles that should continue to be paramount even in the light of future developments. We consider the relative merits of using regular expressions or profiles for SLiM discovery and discuss the main considerations for both predicting new instances of known SLiMs, and de novo prediction of novel SLiMs. In particular, we highlight the importance of correctly modelling evolutionary relationships and the probability of false positive predictions. Fil: Edwards, Richard J.. University of New South Wales; Australia. University of Southampton; Reino Unido Fil: Palopoli, Nicolás. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina. University of Southampton; Reino Unido