Ancient gene duplications in RNA viruses revealed by protein tertiary structure comparisons

To date only a handful of duplicated genes have been described in RNA viruses. This shortage can be attributed to different factors, including the RNA viruses with high mutation rate that would make a large genome more prone to acquire deleterious mutations. This may explain why sequence-based approaches have only found duplications in their most recent evolutionary history. To detect earlier duplications, we performed protein tertiary structure comparisons for every RNA virus family represented in the Protein Data Bank. We present a list of thirty pairs of possible paralogs with <30 per cent sequence identity. It is argued that these pairs are the outcome of six duplication events. These include the α and β subunits of the fungal toxin KP6 present in the dsRNA Ustilago maydis virus (family Totiviridae), the SARS-CoV (Coronaviridae) nsp3 domains SUD-N, SUD-M and X-domain, the Picornavirales (families Picornaviridae, Dicistroviridae, Iflaviridae and Secoviridae) capsid proteins VP1, VP2 and VP3, and the Enterovirus (family Picornaviridae) 3C and 2A cysteine-proteases. Protein tertiary structure comparisons may reveal more duplication events as more three-dimensional protein structures are determined and suggests that, although still rare, gene duplications may be more frequent in RNA viruses than previously thought. Keywords: gene duplications; RNA viruses.

Template-based prediction of protein structure with deep learning

Background Accurate prediction of protein structure is fundamentally important to understand biological function of proteins. Template-based modeling, including protein threading and homology modeling, is a popular method for protein tertiary structure prediction. However, accurate template-query alignment and template selection are still very challenging, especially for the proteins with only distant homologs available. Results We propose a new template-based modelling method called ThreaderAI to improve protein tertiary structure prediction. ThreaderAI formulates the task of aligning query sequence with template as the classical pixel classification problem in computer vision and naturally applies deep residual neural network in prediction. ThreaderAI first employs deep learning to predict residue-residue aligning probability matrix by integrating sequence profile, predicted sequential structural features, and predicted residue-residue contacts, and then builds template-query alignment by applying a dynamic programming algorithm on the probability matrix. We evaluated our methods both in generating accurate template-query alignment and protein threading. Experimental results show that ThreaderAI outperforms currently popular template-based modelling methods HHpred, CNFpred, and the latest contact-assisted method CEthreader, especially on the proteins that do not have close homologs with known structures. In particular, in terms of alignment accuracy measured with TM-score, ThreaderAI outperforms HHpred, CNFpred, and CEthreader by 56, 13, and 11%, respectively, on template-query pairs at the similarity of fold level from SCOPe data. And on CASP13’s TBM-hard data, ThreaderAI outperforms HHpred, CNFpred, and CEthreader by 16, 9 and 8% in terms of TM-score, respectively. Conclusions These results demonstrate that with the help of deep learning, ThreaderAI can significantly improve the accuracy of template-based structure prediction, especially for distant-homology proteins.

A protein tertiary structure mimetic modulator of the Hippo signalling pathway

Transcription factors are key protein effectors in the regulation of gene transcription, and in many cases their activity is regulated via a complex network of protein–protein interactions (PPI). The chemical modulation of transcription factor activity is a long-standing goal in drug discovery but hampered by the difficulties associated with the targeting of PPIs, in particular when extended and flat protein interfaces are involved. Peptidomimetics have been applied to inhibit PPIs, however with variable success, as for certain interfaces the mimicry of a single secondary structure element is insufficient to obtain high binding affinities. Here, we describe the design and characterization of a stabilized protein tertiary structure that acts as an inhibitor of the interaction between the transcription factor TEAD and its co-repressor VGL4, both playing a central role in the Hippo signalling pathway. Modification of the inhibitor with a cell-penetrating entity yielded a cell-permeable proteomimetic that activates cell proliferation via regulation of the Hippo pathway, highlighting the potential of protein tertiary structure mimetics as an emerging class of PPI modulators.

DeepPSC (protein structure camera): computer vision-based protein backbone structure reconstruction from alpha carbon trace as a case study

Deep learning has been increasingly used in protein tertiary structure prediction, a major goal in life science. However, all the algorithms developed so far mostly use protein sequences as input, whereas the vast amount of protein tertiary structure information available in the Protein Data Bank (PDB) database remains largely unused, because of the inherent complexity of 3D data computation. In this study, we propose Protein Structure Camera (PSC) as an approach to convert protein structures into images. As a case study, we developed a deep learning method incorporating PSC (DeepPSC) to reconstruct protein backbone structures from alpha carbon traces. DeepPSC outperformed all the methods currently available for this task. This PSC approach provides a useful tool for protein structure representation, and for the application of deep learning in protein structure prediction and protein engineering.

Template-based prediction of protein structure with deep learning

Accurate prediction of protein structure is fundamentally important to understand biological function of proteins. Template-based modeling, including protein threading and homology modeling, is a popular method for protein tertiary structure prediction. However, accurate template-query alignment and template selection are still very challenging, especially for the proteins with only distant homologs available. We propose a new template-based modelling method called ThreaderAI to improve protein tertiary structure prediction. ThreaderAI formulates the task of aligning query sequence with template as the classical pixel classification problem in computer vision and naturally applies deep residual neural network in prediction. ThreaderAI first employs deep learning to predict residue-residue aligning probability matrix by integrating sequence profile, predicted sequential structural features, and predicted residueresidue contacts, and then builds template-query alignment by applying a dynamic programming algorithm on the probability matrix. We evaluated our methods both in generating accurate template-query alignment and protein threading. Experimental results show that ThreaderAI outperforms currently popular template-based modelling methods HHpred, CNFpred, and the latest contact-assisted method CEthreader, especially on the proteins that do not have close homologs with known structures. In particular, in terms of alignment accuracy measured with TM-score, ThreaderAI outperforms HHpred, CNFpred, and CEthreader by 56%, 13%, and 11%, respectively, on template-query pairs at the similarity of fold level from SCOPe data. And on CASP13’s TBM-hard data, ThreaderAI outperforms HHpred, CNFpred, and CEthreader by 16%, 9% and 8% in terms of TM-score, respectively. These results demonstrate that with the help of deep learning, ThreaderAI can significantly improve the accuracy of template-based structure prediction, especially for distant-homology proteins.Availabilityhttps://github.com/ShenLab/ThreaderAI