Search

Variational autoencoders (VAEs) defined over SMILES string and graph-based representations of molecules promise to improve the optimization of molecular properties, thereby revolutionizing the pharmaceuticals and materials industries. However, these VAEs are hindered by the non-unique nature of SMILES strings and the computational cost of graph convolutions. To efficiently pass messages along all paths through the molecular graph, we encode multiple SMILES strings of a single molecule using a set of stacked recurrent neural networks, pooling hidden representations of each atom between SMILES representations, and use attentional pooling to build a final fixed-length latent representation. By then decoding to a disjoint set of SMILES strings of the molecule, our All SMILES VAE learns an almost bijective mapping between molecules and latent representations near the high-probability-mass subspace of the prior. Our SMILES-derived but molecule-based latent representations significantly surpass the state-of-the-art in a variety of fully- and semi-supervised property regression and molecular property optimization tasks., Comment: Expanded acronym in title

In silico drug-target interaction (DTI) prediction is an important and challenging problem in biomedical research with a huge potential benefit to the pharmaceutical industry and patients. Most existing methods for DTI prediction including deep learning models generally have binary endpoints, which could be an oversimplification of the problem, and those methods are typically unable to handle cold-target problems, i.e., problems involving target protein that never appeared in the training set. Towards this, we contrived PADME (Protein And Drug Molecule interaction prEdiction), a framework based on Deep Neural Networks, to predict real-valued interaction strength between compounds and proteins without requiring feature engineering. PADME takes both compound and protein information as inputs, so it is capable of solving cold-target (and cold-drug) problems. To our knowledge, we are the first to combine Molecular Graph Convolution (MGC) for compound featurization with protein descriptors for DTI prediction. We used multiple cross-validation split schemes and evaluation metrics to measure the performance of PADME on multiple datasets, including the ToxCast dataset, and PADME consistently dominates baseline methods. The results of a case study, which predicts the binding affinity between various compounds and androgen receptor (AR), suggest PADME's potential in drug development. The scalability of PADME is another advantage in the age of Big Data.

The results of the validation of a subset of detected mutations on MiSeq, Illumina. The run was designed to test 23 mutations in 11 cfDNA samples (both amplified and non-amplified). WGA samples are marked with *, n/a â sample not sequenced on MiSeq. Only two calls were not supported on MiSeq. S889G call in VC-012-t1 unamplified cfDNA was not detected on original 454 run, or on MiSeq resequencing. (DOCX 23Â kb)

Background A new sequence independent bioinformatics approach allowing genome-wide search for proteins with similar three dimensional structures has been developed. By utilizing the numerical output of the sequence threading it establishes putative non-obvious structural similarities between proteins. When applied to the testing set of proteins with known three dimensional structures the developed approach was able to recognize structurally similar proteins with high accuracy. Results The method has been developed to identify pathogenic proteins with low sequence identity and high structural similarity to host analogues. Such protein structure relationships would be hypothesized to arise through convergent evolution or through ancient horizontal gene transfer events, now undetectable using current sequence alignment techniques. The pathogen proteins, which could mimic or interfere with host activities, would represent candidate virulence factors. The developed approach utilizes the numerical outputs from the sequence-structure threading. It identifies the potential structural similarity between a pair of proteins by correlating the threading scores of the corresponding two primary sequences against the library of the standard folds. This approach allowed up to 64% sensitivity and 99.9% specificity in distinguishing protein pairs with high structural similarity. Conclusion Preliminary results obtained by comparison of the genomes of Homo sapiens and several strains of Chlamydia trachomatis have demonstrated the potential usefulness of the method in the identification of bacterial proteins with known or potential roles in virulence.

Background Substantial amounts of data on cell signaling, metabolic, gene regulatory and other biological pathways have been accumulated in literature and electronic databases. Conventionally, this information is stored in the form of pathway diagrams and can be characterized as highly "compartmental" (i.e. individual pathways are not connected into more general networks). Current approaches for representing pathways are limited in their capacity to model molecular interactions in their spatial and temporal context. Moreover, the critical knowledge of cause-effect relationships among signaling events is not reflected by most conventional approaches for manipulating pathways. Results We have applied a semantic network (SN) approach to develop and implement a model for cell signaling pathways. The semantic model has mapped biological concepts to a set of semantic agents and relationships, and characterized cell signaling events and their participants in the hierarchical and spatial context. In particular, the available information on the behaviors and interactions of the PI3K enzyme family has been integrated into the SN environment and a cell signaling network in human macrophages has been constructed. A SN-application has been developed to manipulate the locations and the states of molecules and to observe their actions under different biological scenarios. The approach allowed qualitative simulation of cell signaling events involving PI3Ks and identified pathways of molecular interactions that led to known cellular responses as well as other potential responses during bacterial invasions in macrophages. Conclusions We concluded from our results that the semantic network is an effective method to model cell signaling pathways. The semantic model allows proper representation and integration of information on biological structures and their interactions at different levels. The reconstruction of the cell signaling network in the macrophage allowed detailed investigation of connections among various essential molecules and reflected the cause-effect relationships among signaling events. The simulation demonstrated the dynamics of the semantic network, where a change of states on a molecule can alter its function and potentially cause a chain-reaction effect in the system.

Background Using sequence-structure threading we have conducted structural characterization of complete proteomes of 37 archaeal, bacterial and eukaryotic organisms (including worm, fly, mouse and human) totaling 167,888 genes. Results The reported data represent first rather general evaluation of performance of full sequence-structure threading on multiple genomes providing opportunity to evaluate its general applicability for large scale studies. According to the estimated results the sequence-structure threading has assigned protein folds to more then 60% of eukaryotic, 68% of archaeal and 70% of bacterial proteomes. The repertoires of protein classes, architectures, topologies and homologous superfamilies (according to the CATH 2.4 classification) have been established for distant organisms and superkingdoms. It has been found that the average abundance of CATH classes decreases from "alpha and beta" to "mainly beta", followed by "mainly alpha" and "few secondary structures". 3-Layer (aba) Sandwich has been characterized as the most abundant protein architecture and Rossman fold as the most common topology. Conclusion The analysis of genomic occurrences of CATH 2.4 protein homologous superfamilies and topologies has revealed the power-law character of their distributions. The corresponding double logarithmic "frequency – genomic occurrence" dependences characteristic of scale-free systems have been established for individual organisms and for three superkingdoms. Supplementary materials to this works are available at 1.

Background: Using sequence-structure threading we have conducted structural characterization of complete proteomes of 37 archaeal, bacterial and eukaryotic organisms (including worm, fly, mouse and human) totaling 167,888 genes. Results The reported data represent first rather general evaluation of performance of full sequence-structure threading on multiple genomes providing opportunity to evaluate its general applicability for large scale studies. According to the estimated results the sequence-structure threading has assigned protein folds to more then 60% of eukaryotic, 68% of archaeal and 70% of bacterial proteomes. The repertoires of protein classes, architectures, topologies and homologous superfamilies (according to the CATH 2.4 classification) have been established for distant organisms and superkingdoms. It has been found that the average abundance of CATH classes decreases from "alpha and beta" to "mainly beta", followed by "mainly alpha" and "few secondary structures". 3-Layer (aba) Sandwich has been characterized as the most abundant protein architecture and Rossman fold as the most common topology. Conclusion The analysis of genomic occurrences of CATH 2.4 protein homologous superfamilies and topologies has revealed the power-law character of their distributions. The corresponding double logarithmic "frequency – genomic occurrence" dependences characteristic of scale-free systems have been established for individual organisms and for three superkingdoms. Supplementary materials to this works are available at [1].

Background We establish that the occurrence of protein folds among genomes can be accurately described with a Weibull function. Systems which exhibit Weibull character can be interpreted with reliability theory commonly used in engineering analysis. For instance, Weibull distributions are widely used in reliability, maintainability and safety work to model time-to-failure of mechanical devices, mechanisms, building constructions and equipment. Results We have found that the Weibull function describes protein fold distribution within and among genomes more accurately than conventional power functions which have been used in a number of structural genomic studies reported to date. It has also been found that the Weibull reliability parameter β for protein fold distributions varies between genomes and may reflect differences in rates of gene duplication in evolutionary history of organisms. Conclusions The results of this work demonstrate that reliability analysis can provide useful insights and testable predictions in the fields of comparative and structural genomics.

Background. A new sequence independent bioinformatics approach allowing genome-wide search for proteins with similar three dimensional structures has been developed. By utilizing the numerical output of the sequence threading it establishes putative non-obvious structural similarities between proteins. When applied to the testing set of proteins with known three dimensional structures the developed approach was able to recognize structurally similar proteins with high accuracy. Results The method has been developed to identify pathogenic proteins with low sequence identity and high structural similarity to host analogues. Such protein structure relationships would be hypothesized to arise through convergent evolution or through ancient horizontal gene transfer events, now undetectable using current sequence alignment techniques. The pathogen proteins, which could mimic or interfere with host activities, would represent candidate virulence factors. The developed approach utilizes the numerical outputs from the sequence-structure threading. It identifies the potential structural similarity between a pair of proteins by correlating the threading scores of the corresponding two primary sequences against the library of the standard folds. This approach allowed up to 64% sensitivity and 99.9% specificity in distinguishing protein pairs with high structural similarity. Conclusion Preliminary results obtained by comparison of the genomes ofHomo sapiensand several strains ofChlamydia trachomatishave demonstrated the potential usefulness of the method in the identification of bacterial proteins with known or potential roles in virulence.

Background: The androgen receptor (AR) is a pivotal drug target for the treatment of prostate cancer, including its lethal castration-resistant (CRPC) form. All current non-steroidal AR antagonists, such as hydroxyflutamide, bicalutamide, and enzalutamide, target the androgen binding site of the receptor, competing with endogenous androgenic steroids. Several AR mutations in this binding site have been associated with poor prognosis and resistance to conventional prostate cancer drugs. In order to develop an effective CRPC therapy, it is crucial to understand the effects of these mutations on the functionality of the AR and its ability to interact with endogenous steroids and conventional AR inhibitors. Results: We previously utilized circulating cell-free DNA (cfDNA) sequencing technology to examine the AR gene for the presence of mutations in CRPC patients. By modifying our sequencing and data analysis approaches, we identify four additional single AR mutations and five mutation combinations associated with CRPC. Importantly, we conduct experimental functionalization of all the AR mutations identified by the current and previous cfDNA sequencing to reveal novel gain-of-function scenarios. Finally, we evaluate the effect of a novel class of AR inhibitors targeting the binding function 3 (BF3) site on the activity of CRPC-associated AR mutants. Conclusions: This work demonstrates the feasibility of a prognostic and/or diagnostic platform combining the direct identification of AR mutants from patients’ serum, and the functional characterization of these mutants in order to provide personalized recommendations regarding the best future therapy.

Recurrent, metastatic prostate cancer continues to be a leading cause of cancer-death in men. The androgen receptor (AR) is a modular, ligand-inducible transcription factor that regulates the expression of genes that can drive the progression of this disease, and as a consequence, this receptor is a key therapeutic target for controlling prostate cancer. The current drugs designed to directly inhibit the AR are called anti-androgens, and all act by competing with androgens for binding to the androgen/ligand binding site. Unfortunately, with the inevitable progression of the cancer to castration resistance, many of these drugs become ineffective. However, there are numerous other regulatory sites on this protein that have not been exploited therapeutically. The regulation of AR activity involves a cascade of complex interactions with numerous chaperones, co-factors and co-regulatory proteins, leading ultimately to direct binding of AR dimers to specific DNA androgen response elements within the promoter and enhancers of androgen-regulated genes. As part of the family of nuclear receptors, the AR is organized into modular structural and functional domains with specialized roles in facilitating their inter-molecular interactions. These regions of the AR present attractive, yet largely unexploited, drug target sites for reducing or eliminating androgen signaling in prostate cancers. The design of small molecule inhibitors targeting these specific AR domains is only now being realized and is the culmination of decades of work, including crystallographic and biochemistry approaches to map the shape and accessibility of the AR surfaces and cavities. Here, we review the structure of the AR protein and describe recent advancements in inhibiting its activity with small molecules specifically designed to target areas distinct from the receptor’s androgen binding site. It is anticipated that these new classes of anti-AR drugs will provide an additional arsenal to treat castration-resistant prostate cancer.

Cutaneous T-cell lymphomas (CTCL) are the most common primary lymphomas of the skin. We have previously identified thymocyte selection-associated high mobility group (HMG) box protein (TOX) as a promising drug target in CTCL; however, there are currently no small molecules able to directly inhibit TOX. We aimed to address this unmet opportunity by developing anti-TOX therapeutics with the use of computer-aided drug discovery methods. The available NMR-resolved structure of the TOX protein was used to model its DNA-binding HMG-box domain. To investigate the druggability of the corresponding protein–DNA interface on TOX, we performed a pilot virtual screening of 200,000 small molecules using in silico docking and identified ‘hot spots’ for drug-binding on the HMG-box domain. We then performed a large-scale virtual screening of 7.6 million drug-like compounds that were available from the ZINC15 database. As a result, a total of 140 top candidate compounds were selected for subsequent in vitro validation. Of those, 18 small molecules have been characterized as selective TOX inhibitors.

The response of the CRPC-associated mutants to increasing concentrations of anti-androgens. Abiraterone, bicalutamide, hydroxyflutamide, enzalutamide, ARN509, and an in-house developed AR inhibitor VPC-13566 were tested against the 24 CRPC-associated AR mutants. Each concentration was assayed in quadruplicate n = 4, with a biological replicate of n = 3. Results were averaged and normalized by expressing them as a percentage of WT AR activity ± SEM. (PDF 345 kb)

The inhibition of the androgen receptor (AR) is an established strategy in prostate cancer (PCa) treatment until drug resistance develops either through mutations in the ligand-binding domain (LBD) portion of the receptor or its deletion. We previously identified a druggable pocket on the DNA binding domain (DBD) dimerization surface of the AR and reported several potent inhibitors that effectively disrupted DBD-DBD interactions and consequently demonstrated certain antineoplastic activity. Here we describe further development of small molecule inhibitors of AR DBD dimerization and provide their broad biological characterization. The developed compounds demonstrate improved activity in the mammalian two-hybrid assay, enhanced inhibition of AR-V7 transcriptional activity, and improved microsomal stability. These findings position us for the development of AR inhibitors with entirely novel mechanisms of action that would bypass most forms of PCa treatment resistance, including the truncation of the LBD of the AR.

Orphan nuclear receptor TLX (NR2E1) plays a critical role in the regulation of neural stem cells (NSC) as well as in the development of NSC-derived brain tumors. In the last years, new data have emerged implicating TLX in prostate and breast cancer. Therefore, inhibitors of TLX transcriptional activity may have a significant impact on the treatment of several critical malignancies. However, the TLX protein possesses a non-canonical ligand-binding domain (LBD), which lacks a ligand-binding pocket (conventionally targeted in case of nuclear receptors) that complicates the development of small molecule inhibitors of TLX. Herein, we utilized a rational structure-based design approach to identify small molecules targeting the Atro-box binding site of human TLX LBD. As a result of virtual screening of ~7 million molecular structures, 97 compounds were identified and evaluated in the TLX-responsive luciferase reporter assay. Among those, three chemicals demonstrated 40–50% inhibition of luciferase-detected transcriptional activity of the TLX orphan nuclear receptor at a dose of 35 µM. The identified compounds represent the first class of small molecule inhibitors of TLX transcriptional activity identified via methods of computer-aided drug discovery.