MILCDock: Machine Learning Enhanced Consensus Docking for Virtual Screening in Drug Discovery.

Molecular docking tools are regularly used to computationally identify new molecules in virtual screening for drug discovery. However, docking tools suffer from inaccurate scoring functions with widely varying performance on different proteins. To enable more accurate ranking of active over inactive ligands in virtual screening, we created a machine learning consensus docking tool, MILCDock, that uses predictions from five traditional molecular docking tools to predict the probability a ligand binds to a protein. MILCDock was trained and tested on data from both the DUD-E and LIT-PCBA docking datasets and shows improved performance over traditional molecular docking tools and other consensus docking methods on the DUD-E dataset. LIT-PCBA targets proved to be difficult for all methods tested. We also find that DUD-E data, although biased, can be effective in training machine learning tools if care is taken to avoid DUD-E's biases during training.

Detecting selection in immunoglobulin sequences.

The ability to detect selection by analyzing mutation patterns in experimentally derived immunoglobulin (Ig) sequences is a critical part of many studies. Such techniques are useful not only for understanding the response to pathogens, but also to determine the role of antigen-driven selection in autoimmunity, B cell cancers and the diversification of pre-immune repertoires in certain species. Despite its importance, quantifying selection in experimentally derived sequences is fraught with difficulties. The necessary parameters for statistical tests (such as the expected frequency of replacement mutations in the absence of selection) are non-trivial to calculate, and results are not easily interpretable when analyzing more than a handful of sequences. We have developed a web server that implements our previously proposed Focused binomial test for detecting selection. Several features are integrated into the web site in order to facilitate analysis, including V(D)J germline segment identification with IMGT alignment, batch submission of sequences and integration of additional test statistics proposed by other groups. We also implement a Z-score-based statistic that increases the power of detecting selection while maintaining specificity, and further allows for the combined analysis of sequences from different germlines. The tool is freely available at http://clip.med.yale.edu/selection.

A probabilistic view of protein stability, conformational specificity, and design.

Various approaches have used neural networks as probabilistic models for the design of protein sequences. These "inverse folding" models employ different objective functions, which come with trade-offs that have not been assessed in detail before. This study introduces probabilistic definitions of protein stability and conformational specificity and demonstrates the relationship between these chemical properties and the [Formula: see text] Boltzmann probability objective. This links the Boltzmann probability objective function to experimentally verifiable outcomes. We propose a novel sequence decoding algorithm, referred to as "BayesDesign", that leverages Bayes' Rule to maximize the [Formula: see text] objective instead of the [Formula: see text] objective common in inverse folding models. The efficacy of BayesDesign is evaluated in the context of two protein model systems, the NanoLuc enzyme and the WW structural motif. Both BayesDesign and the baseline ProteinMPNN algorithm increase the thermostability of NanoLuc and increase the conformational specificity of WW. The possible sources of error in the model are analyzed.