Biology subjects

Roberts, T. J.

Publications and source records attributed to Roberts, T. J..

3 recordsLinked to original sources

bioRxiv2020

VAE-Sim: a novel molecular similarity measure based on a variational autoencoder

Molecular similarity is an elusive but core unsupervised cheminformatics concept, yet different fingerprint encodings of molecular structures return very different similarity values even when using the same similarity metric. Each encoding may be of value when applied to other problems with objective or target functions, implying that a priori none is better than the others, nor than encoding-free metrics such as maximum common substructure (MCSS). We here introduce a novel approach to molecular similarity, in the form of a variational autoencoder (VAE). This learns the joint distribution p(z|x) where z is a latent vector and x are the (same) input/output data. It takes the form of a bowtie-shaped artificial neural network. In the middle is a bottleneck layer or latent vector in which inputs are transformed into, and represented as, a vector of numbers (encoding), with a reverse process (decoding) seeking to return the SMILES string that was the input. We train a VAE on over 6 million druglike molecules and natural products (including over one million in the final holdout set). The VAE vector distances provide a rapid and novel metric for molecular similarity that is both easily and rapidly calculated. We describe the method and its application to a typical similarity problem in cheminformatics.

pharmacology and toxicology↗

bioRxiv2020

The Atypical Fibrin Fibre Network in Rheumatoid Arthritis and its Relation to Autoimmunity, Inflammation and Thrombosis

ObjectiveThe risk of cardiovascular events in patients with RA is disproportionately heightened as a result of systemic inflammation. The relative effect of autoimmune-associated citrullination on the structure and thrombotic potential of fibrin(ogen) remains unknown. We therefore compared indices of vascular function, inflammation, coagulation and fibrin clot composition in RA patients with healthy controls and evaluated inter-parameter relationships. MethodsBlood samples were collected from 30 RA patients and 25 age- and gender-matched healthy volunteers. Levels of SAA, CRP, ICAM-1 and VCAM-1 was measured using a sandwich immunoassay. Whole blood coagulation was assessed using Thromboelastography. Fibrin clot networks and fiber structure was investigated using Scanning Electron Microscopy. The detection and quantification of citrullination in formed fibrin clots were performed using a fluorescently labeled Citrulline monoclonal antibody with Confocal Microscopy. ResultsConcentrations of SAA, CRP and ICAM-1 were significantly elevated in RA patients compared to controls. TEG parameters relating to coagulation initiation (R and K), rate of fibrin cross-linking (-Angle), and time to reach maximum thrombus generation (TMRTG) were attenuated in RA patients. Parameters relating to clot strength (MA, MRTG, TGG) did not statistically differ between RA and controls. Logistic regression modelling revealed stronger association between acute phase reactants (CRP, SAA) with TEG parameters than endothelial function markers. Microscopic analysis revealed denser networks of thicker fibrin fibers in RA patients compared to controls [median (interquartile range) 214 (170-285) vs 120 (100-144) nm respectively, p<0.0001, Odds ratio=22.7). Detection of multiple citrullinated regions within fibrin clot structures in RA patients, which was less prevalent in control samples (p<0.05, OR=2.2). ConclusionPatients with active RA display a coagulation profile that is dissimilar to general findings associated with other inflammatory conditions. The alteration of protein structures by autoimmune linked citrullination could play a role in determining the structure of fibrin and the potential of conferring a heightened thrombotic risk in RA patients.

biochemistry↗

bioRxiv2020

DeepGraphMol, a multi-objective, computational strategy for generating molecules with desirable properties: a graph convolution and reinforcement learning approach

We address the problem of generating novel molecules with desired interaction properties as a multi-objective optimization problem. Interaction binding models are learned from binding data using graph convolution networks (GCNs). Since the experimentally obtained property scores are recognised as having potentially gross errors, we adopted a robust loss for the model. Combinations of these terms, including drug likeness and synthetic accessibility, are then optimized using reinforcement learning based on a graph convolution policy approach. Some of the molecules generated, while legitimate chemically, can have excellent drug-likeness scores but appear unusual. We provide an example based on the binding potency of small molecules to dopamine transporters. We extend our method successfully to use a multi-objective reward function, in this case for generating novel molecules that bind with dopamine transporters but not with those for norepinephrine. Our method should be generally applicable to the generation in silico of molecules with desirable properties.

bioinformatics↗