Predicting the biological function of molecules, be it proteins or drug-like compounds, from their atomic structure is an important and long-standing problem. The electron density field and electrostatic potential field of a molecule contain the “raw fingerprint” of how this molecule can fit to binding partners. In this paper, we show that deep learning can predict biological function of molecules directly from their raw 3D approximated electron density and electrostatic potential fields. Protein function based on Enzyme Commission numbers is predicted from the approximated electron density field. In another experiment, the activity of small molecules is predicted with quality comparable to state-of-the-art descriptor-based methods. We propose several alternative computational models for the GPU with different memory and runtime requirements for different sizes of molecules and of databases. We also propose application-specific multi-channel data representations.

MOTIVATION: A few years ago it was shown that by using a maximum entropy approach to describe couplings between columns in a multiple sequence alignment it is possible to significantly increase the accuracy of residue contact predictions. For very large protein families with more than 1000 effective sequences the accuracy is sufficient to produce accurate models of proteins as well as complexes. Today, for about half of all Pfam domain families no structure is known, but unfortunately most of these families have at most a few hundred members, i.e. are too small for such contact prediction methods. RESULTS: To extend accurate contact predictions to the thousands of smaller protein families we present PconsC3, a fast and improved method for protein contact predictions that can be used for families with even 100 effective sequence members. PconsC3 outperforms direct coupling analysis (DCA) methods significantly independent on family size, secondary structure content, contact range, or the number of selected contacts. AVAILABILITY AND IMPLEMENTATION: PconsC3 is available as a web server and downloadable version at http://c3.pcons.net . The downloadable version is free for all to use and licensed under the GNU General Public License, version 2. At this site contact predictions for most Pfam families are also available. We do estimate that more than 4000 contact maps for Pfam families of unknown structure have more than 50% of the top-ranked contacts predicted correctly. CONTACT: [email protected]. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Proteins are the "building blocks of life", the most abundant organic molecules, and the central focus of most areas of biomedicine. Protein structure is strongly related to protein function, thus structure prediction is a crucial task on the way to solve many biological questions. A contact map is a compact representation of the three-dimensional structure of a protein via the pairwise contacts between the amino acid constituting the protein. We use a convolutional network to calculate protein contact maps from inferred statistical coupling between positions in the protein sequence. The input to the network has an image-like structure amenable to convolutions, but every "pixel" instead of color channels contains a bipartite undirected edge-weighted graph. We propose several methods for treating such "graph-valued images" in a convolutional network. The proposed method outperforms state-of-the-art methods by a large margin. It also allows for a great flexibility with regard to the input data, which makes it useful for studying a wide range of problems.

MOTIVATION: Recently it has been shown that the quality of protein contact prediction from evolutionary information can be improved significantly if direct and indirect information is separated. Given sufficiently large protein families, the contact predictions contain sufficient information to predict the structure of many protein families. However, since the first studies contact prediction methods have improved. Here, we ask how much the final models are improved if improved contact predictions are used. RESULTS: In a small benchmark of 15 proteins, we show that the TM-scores of top-ranked models are improved by on average 33% using PconsFold compared with the original version of EVfold. In a larger benchmark, we find that the quality is improved with 15-30% when using PconsC in comparison with earlier contact prediction methods. Further, using Rosetta instead of CNS does not significantly improve global model accuracy, but the chemistry of models generated with Rosetta is improved. AVAILABILITY: PconsFold is a fully automated pipeline for ab initio protein structure prediction based on evolutionary information. PconsFold is based on PconsC contact prediction and uses the Rosetta folding protocol. Due to its modularity, the contact prediction tool can be easily exchanged. The source code of PconsFold is available on GitHub at https://www.github.com/ElofssonLab/pcons-fold under the MIT license. PconsC is available from http://c.pcons.net/. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Correlation patterns in multiple sequence alignments of homologous proteins can be exploited to infer information on the three-dimensional structure of their members. The typical pipeline to address this task, which we in this paper refer to as the three dimensions of contact prediction, is to (i) filter and align the raw sequence data representing the evolutionarily related proteins; (ii) choose a predictive model to describe a sequence alignment; (iii) infer the model parameters and interpret them in terms of structural properties, such as an accurate contact map. We show here that all three dimensions are important for overall prediction success. In particular, we show that it is possible to improve significantly along the second dimension by going beyond the pair-wise Potts models from statistical physics, which have hitherto been the focus of the field. These (simple) extensions are motivated by multiple sequence alignments often containing long stretches of gaps which, as a data feature, would be rather untypical for independent samples drawn from a Potts model. Using a large test set of proteins we show that the combined improvements along the three dimensions are as large as any reported to date.

Given sufficient large protein families, and using a global statistical inference approach, it is possible to obtain sufficient accuracy in protein residue contact predictions to predict the structure of many proteins. However, these approaches do not consider the fact that the contacts in a protein are neither randomly, nor independently distributed, but actually follow precise rules governed by the structure of the protein and thus are interdependent. Here, we present PconsC2, a novel method that uses a deep learning approach to identify protein-like contact patterns to improve contact predictions. A substantial enhancement can be seen for all contacts independently on the number of aligned sequences, residue separation or secondary structure type, but is largest for β-sheet containing proteins. In addition to being superior to earlier methods based on statistical inferences, in comparison to state of the art methods using machine learning, PconsC2 is superior for families with more than 100 effective sequence homologs. The improved contact prediction enables improved structure prediction.

SUMMARY: Recently, several new contact prediction methods have been published. They use (i) large sets of multiple aligned sequences and (ii) assume that correlations between columns in these alignments can be the results of indirect interaction. These methods are clearly superior to earlier methods when it comes to predicting contacts in proteins. Here, we demonstrate that combining predictions from two prediction methods, PSICOV and plmDCA, and two alignment methods, HHblits and jackhmmer at four different e-value cut-offs, provides a relative improvement of 20% in comparison with the best single method, exceeding 70% correct predictions for one contact prediction per residue. AVAILABILITY: The source code for PconsC along with supplementary data is freely available at http://c.pcons.net/ CONTACT: [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

SUMMARY: Clustering methods are often needed for accurately assessing the quality of modeled protein structures. Recent blind evaluation of quality assessment methods in CASP10 showed that there is little difference between many different methods as far as ranking models and selecting best model are concerned. When comparing many models, the computational cost of the model comparison can become significant. Here, we present PconsD, a fast, stream-computing method for distance-driven model quality assessment that runs on consumer hardware. PconsD is at least one order of magnitude faster than other methods of comparable accuracy. AVAILABILITY: The source code for PconsD is freely available at http://d.pcons.net/. Supplementary benchmarking data are also available there. CONTACT: [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

UNLABELLED: Multiple templates can often be used to build more accurate homology models than models built from a single template. Here we introduce PconsM, an automated protocol that uses multiple templates to build protein models. PconsM has been among the top-performing methods in the recent CASP experiments and consistently perform better than the single template models used in Pcons.net. In particular for the easier targets with many alternative templates with a high degree of sequence identity, quality is readily improved with a few percentages over the highest ranked model built on a single template. PconsM is available as an additional pipeline within the Pcons.net protein structure prediction server. AVAILABILITY AND IMPLEMENTATION: PconsM is freely available from http://pcons.net/.

SUMMARY: SPOCTOPUS is a method for combined prediction of signal peptides and membrane protein topology, suitable for genome-scale studies. Its objective is to minimize false predictions of transmembrane regions as signal peptides and vice versa. We provide a description of the SPOCTOPUS algorithm together with a performance evaluation where SPOCTOPUS compares favorably with state-of-the-art methods for signal peptide and topology predictions. AVAILABILITY: SPOCTOPUS is available as a web server and both the source code and benchmark data are available for download at http://octopus.cbr.su.se/