Search

Development of novel anti-cancer treatments requires not only a comprehensive knowledge of cancer processes and drug mechanisms of action, but also the ability to accurately predict the response of various cancer cell lines to therapeutics. Numerous computational methods have been developed to address this issue, including algorithms employing supervised machine learning. Nonetheless, high prediction accuracies reported for many of these techniques may result from a significant overlap among training, validation, and testing sets, making existing predictors inapplicable to new data. To address these issues, we developed CancerOmicsNet, a graph neural network with sophisticated attention propagation mechanisms to predict the therapeutic effects of kinase inhibitors across various tumors. Emphasizing on the system-level complexity of cancer, CancerOmicsNet integrates multiple heterogeneous data, such as biological networks, genomics, inhibitor profiling, and gene-disease associations, into a unified graph structure. The performance of CancerOmicsNet, properly cross-validated at the tissue level, is 0.83 in terms of the area under the receiver operating characteristics, which is notably higher than those measured for other approaches. CancerOmicsNet generalizes well to unseen data, i.e., it can predict therapeutic effects across a variety of cancer cell lines and inhibitors. CancerOmicsNet is freely available to the academic community at https://github.com/pulimeng/CancerOmicsNet.

Computational modeling is an essential component of modern drug discovery. One of its most important applications is to select promising drug candidates for pharmacologically relevant target proteins. Because of continuing advances in structural biology, putative binding sites for small organic molecules are being discovered in numerous proteins linked to various diseases. These valuable data offer new opportunities to build efficient computational models predicting binding molecules for target sites through the application of data mining and machine learning. In particular, deep neural networks are powerful techniques capable of learning from complex data in order to make informed drug binding predictions. In this communication, we describe Pocket2Drug, a deep graph neural network model to predict binding molecules for a given a ligand binding site. This approach first learns the conditional probability distribution of small molecules from a large dataset of pocket structures with supervised training, followed by the sampling of drug candidates from the trained model. Comprehensive benchmarking simulations show that using Pocket2Drug significantly improves the chances of finding molecules binding to target pockets compared to traditional drug selection procedures. Specifically, known binders are generated for as many as 80.5% of targets present in the testing set consisting of dissimilar data from that used to train the deep graph neural network model. Overall, Pocket2Drug is a promising computational approach to inform the discovery of novel biopharmaceuticals.

Traditional techniques to identify macromolecular targets for drugs utilize solely the information on a query drug and a putative target. Nonetheless, the mechanisms of action of many drugs depend not only on their binding affinity toward a single protein, but also on the signal transduction through cascades of molecular interactions leading to certain phenotypes. Although using protein-protein interaction networks and drug-perturbed gene expression profiles can facilitate system-level investigations of drug-target interactions, utilizing such large and heterogeneous data poses notable challenges. To improve the state-of-the-art in drug target identification, we developed GraphDTI, a robust machine learning framework integrating the molecular-level information on drugs, proteins, and binding sites with the system-level information on gene expression and protein-protein interactions. In order to properly evaluate the performance of GraphDTI, we compiled a high-quality benchmarking dataset and devised a new cluster-based cross-validation protocol. Encouragingly, GraphDTI not only yields an AUC of 0.996 against the validation dataset, but it also generalizes well to unseen data with an AUC of 0.939, significantly outperforming other predictors. Finally, selected examples of identified drugtarget interactions are validated against the biomedical literature. Numerous applications of GraphDTI include the investigation of drug polypharmacological effects, side effects through offtarget binding, and repositioning opportunities.

Advances in several engineering fields have led to a trend toward miniaturization and portability of wearable biosensing devices, which used to be confined to large tools and clinical settings. Various systems to continuously measure electrophysiological activity through electrical and optical methods are one category of such devices. Being wearable and intended for prolonged use, the amount of noise introduced on sensors by movement remains a challenge and requires further optimization. User movement causes motion artifacts that alter the overall quality of the signals obtained, hence corrupting the resulting measurements. This paper introduces a fully wearable optical biosensing system to continuously measure pulse oximetry and heart rate, utilizing a reflectance-based probe. Furthermore, a novel data-dependent motion artifact tailoring algorithm is implemented to eliminate noisy data due to the motion artifact and measure oxygenation level with high accuracy in real time. By taking advantages of current wireless transmission and signal processing technologies, the developed wearable photoplethysmography device successfully captures the measured signals and sends them wirelessly to a mobile device for signal processing in real time. After applying motion artifact tailoring, evaluating accuracy with a continuous clinical device, the blood oxygenation measurements obtained from our system yielded an accuracy of at least 98%, when compared to a range of 93.6%-96.7% observed before from the same initial data. Additionally, heart rate accuracy above 97% was achieved. Motion artifact tailoring and removal in real time, continuous systems will allow wearable devices to be truly wearable and a reliable electrophysiological monitoring and diagnostics tool for everyday use.

The efficiency of drug development defined as a number of successfully launched new pharmaceuticals normalized by financial investments has significantly declined. Nonetheless, recent advances in high-throughput experimental techniques and computational modeling promise reductions in the costs and development times required to bring new drugs to market. The prediction of toxicity of drug candidates is one of the important components of modern drug discovery. In this work, we describe eToxPred, a new approach to reliably estimate the toxicity and synthetic accessibility of small organic compounds. eToxPred employs machine learning algorithms trained on molecular fingerprints to evaluate drug candidates. The performance is assessed against multiple datasets containing known drugs, potentially hazardous chemicals, natural products, and synthetic bioactive compounds. Encouragingly, eToxPred predicts the synthetic accessibility with the mean square error of only 4% and the toxicity with the accuracy of as high as 72%. eToxPred can be incorporated into protocols to construct custom libraries for virtual screening in order to filter out those drug candidates that are potentially toxic or would be difficult to synthesize. It is freely available as a stand-alone software at https://github.com/pulimeng/etoxpred .

Genomic profiles of cancer cells provide valuable information on genetic alterations in cancer. Several recent studies employed these data to predict the response of cancer cell lines to treatment with drugs. Nonetheless, due to the multifactorial phenotypes and intricate mechanisms of cancer, the accurate prediction of the effect of pharmacotherapy on a specific cell line based on the genetic information alone is problematic. High prediction accuracies reported in the literature likely result from significant overlaps among training, validation, and testing sets, making many predictors inapplicable to new data. To address these issues, we developed GraphGR, a graph neural network with sophisticated attention propagation mechanisms to predict the therapeutic effects of kinase inhibitors across various tumors. Emphasizing on the system-level complexity of cancer, GraphGR integrates multiple heterogeneous data, such as biological networks, genomics, inhibitor profiling, and genedisease associations, into a unified graph structure. In order to construct diverse and information-rich cancer-specific networks, we devised a novel graph reduction protocol based on not only the topological information, but also the biological knowledge. The performance of GraphGR, properly cross-validated at the tissue level, is 0.83 in terms of the area under the receiver operating characteristics, which is notably higher than those measured for other approaches on the same data. Finally, several new predictions are validated against the biomedical literature demonstrating that GraphGR generalizes well to unseen data, i.e. it can predict therapeutic effects across a variety of cancer cell lines and inhibitors. GraphGR is freely available to the academic community at https://github.com/pulimeng/GraphGR.

Motivation Fast and accurate classification of ligand-binding sites in proteins with respect to the class of binding molecules is invaluable not only to the automatic functional annotation of large datasets of protein structures but also to projects in protein evolution, protein engineering and drug development. Deep learning techniques, which have already been successfully applied to address challenging problems across various fields, are inherently suitable to classify ligand-binding pockets. Our goal is to demonstrate that off-the-shelf deep learning models can be employed with minimum development effort to recognize nucleotide- and heme-binding sites with a comparable accuracy to highly specialized, voxel-based methods. Results We developed BionoiNet, a new deep learning-based framework implementing a popular ResNet model for image classification. BionoiNet first transforms the molecular structures of ligand-binding sites to 2D Voronoi diagrams, which are then used as the input to a pretrained convolutional neural network classifier. The ResNet model generalizes well to unseen data achieving the accuracy of 85.6% for nucleotide- and 91.3% for heme-binding pockets. BionoiNet also computes significance scores of pocket atoms, called BionoiScores, to provide meaningful insights into their interactions with ligand molecules. BionoiNet is a lightweight alternative to computationally expensive 3D architectures. Availability and implementation BionoiNet is implemented in Python with the source code freely available at: https://github.com/CSBG-LSU/BionoiNet. Supplementary information Supplementary data are available at Bioinformatics online.

We consider the use of multiple mobile agents to explore an unknown area. The area is orthogonal, such that all perimeter lines run both vertically and horizontally. The area may consist of unknown rectangular holes which are non-traversable internally. For the sake of analysis, we assume that the area is discretized into N points allowing the agents to move from one point to an adjacent one. Mobile agents communicate through face-to-face communication when in adjacent points. The objective of exploration is to develop an online algorithm that will explore the entire area while reducing the total work of all k agents, where the work is measured as the number of points traversed. We propose splitting the exploration into two alternating tasks, perimeter and room exploration. The agents all begin with the perimeter scan and when a room is found they transition to room scan after which they continue with perimeter scan until the next room is found and so on. Given the total traversable points N, our algorithm completes in total O ( N ) work with each agent performing O ( N / k ) work, namely the work is balanced. If the rooms are hole-free the exploration time is also asymptotically optimal, O ( N / k ) . To our knowledge, this is the first agent coordination algorithm that considers simultaneously work balancing and small exploration time.

Comprehensive characterization of ligand-binding sites is invaluable to infer molecular functions of hypothetical proteins, trace evolutionary relationships between proteins, engineer enzymes to achieve a desired substrate specificity, and develop drugs with improved selectivity profiles. These research efforts pose significant challenges owing to the fact that similar pockets are commonly observed across different folds, leading to the high degree of promiscuity of ligand-protein interactions at the system-level. On that account, novel algorithms to accurately classify binding sites are needed. Deep learning is attracting a significant attention due to its successful applications in a wide range of disciplines. In this communication, we present DeepDrug3D, a new approach to characterize and classify binding pockets in proteins with deep learning. It employs a state-of-the-art convolutional neural network in which biomolecular structures are represented as voxels assigned interaction energy-based attributes. The current implementation of DeepDrug3D, trained to detect and classify nucleotide- and heme-binding sites, not only achieves a high accuracy of 95%, but also has the ability to generalize to unseen data as demonstrated for steroid-binding proteins and peptidase enzymes. Interestingly, the analysis of strongly discriminative regions of binding pockets reveals that this high classification accuracy arises from learning the patterns of specific molecular interactions, such as hydrogen bonds, aromatic and hydrophobic contacts. DeepDrug3D is available as an open-source program at https://github.com/pulimeng/DeepDrug3D with the accompanying TOUGH-C1 benchmarking dataset accessible from https://osf.io/enz69/., Author summary Small organic ligands bind to the locations of chemical specificity and affinity on their protein targets, called binding sites. A typical ligand-binding site is a small pocket formed by a few residues while the remaining protein structure acts as a framework providing the correct orientation of binding residues. Annotating ligand-binding sites is complicated by a fact that the same small molecule often binds to similar pockets but located in different proteins. In order to improve the detection and classification of binding pockets in proteins, we developed a new computational tool, DeepDrug3D. Our algorithm employs a convolutional neural network, a class of deep learning already commonly used in visual imagery analysis, recommender systems, and natural language processing. DeepDrug3D is able to accurately classify binding sites by learning the patterns of specific molecular interactions between ligands and their protein targets, such as hydrogen bonds, aromatic and hydrophobic contacts. Although the current proof-of-concept implementation is limited to a few most abundant functional classes, the repertoire of pocket types handled by DeepDrug3D will significantly be expanded in the near future.

Photoplethysmogram (PPG) signals are widely used for wearable electronic devices nowadays. The PPG signal is extremely sensitive to the motion artifacts (MAs) caused by the subject's movement. The detection and removal of such MAs remains a difficult problem. Due to the complicated MA signal waveforms, none of the existing techniques can lead to satisfactory results. In this paper, a new framework to identify and tailor the abrupt MAs in PPG is proposed, which consists of feature extraction, change-point detection, and MA removal. In order to achieve the optimal performance, a data-dependent frame-size determination mechanism is employed. Experiments for the heart-beat-rate-measurement application have been conducted to demonstrate the effectiveness of our proposed method, by a correct detection rate of MAs at 98% and the average heart-beat-rate tracking accuracy above 97%. On the other hand, this new framework maintains the original signal temporal structure unlike the spectrum-based approach, and it can be further applied for the calculation of blood oxygen level (SpO2).