Biology subjects

Mangione, W.

Publications and source records attributed to Mangione, W..

12 recordsLinked to original sources

bioRxiv2026

Proteome scale drug prioritization from AlphaFold 3 interaction confidence

AlphaFold 3 (AF3) predicts protein ligand complexes, but the utility of its confidence scores for drug prioritization remains unclear. We benchmarked AF3 interaction scores against BANDOCK+, an experimentally informed approach, using approved drugs and human protein libraries. AF3 ipTM achieved similar aggregate disease ranking performance to BANDOCK+, with differences among individual indications. These results support the evaluation of AF3 interaction confidence as a source of information for proteome scale drug prioritization.

bioinformatics↗

bioRxiv2025

Multiscale analysis and optimal glioma therapeutic candidate discovery using the CANDO platform

Glioma is a highly malignant brain tumor with limited treatment options. We employed the Computational Analysis of Novel Drug Opportunities (CANDO) platform for multiscale therapeutic discovery to predict new glioma therapies. We began by computing interaction scores between extensive libraries of drugs/compounds and proteins to generate "interaction signatures" that model compound behavior on a proteomic scale. Compounds with signatures most similar to those of drugs approved for a given indication were considered potential treatments. These compounds were further ranked by degree of consensus in corresponding similarity lists. We benchmarked performance by measuring the recovery of approved drugs in these similarity and consensus lists at various cutoffs, using multiple metrics and comparing results to random controls and performance across all indications. Compounds ranked highly by consensus but not previously associated with the indication of interest were considered new predictions. Our benchmarking results showed that CANDO improved accuracy in identifying glioma-associated drugs across all cutoffs compared to random controls. Our predictions, supported by literature-based analysis, identified 23 potential glioma treatments, including approved drugs like vitamin D, taxanes, vinca alkaloids, topoisomerase inhibitors, and folic acid, as well as investigational compounds such as ginsenosides, chrysin, resiniferatoxin, and cryptotanshinone. Further functional annotation-based analysis of the top targets with the strongest interactions to these predictions identified Vitamin D3 receptor, thyroid hormone receptor, acetylcholinesterase, cyclin-dependent kinase 2, tubulin alpha chain, dihydrofolate reductase, and thymidylate synthase. These findings indicate that CANDOs multitarget, multiscale framework is effective in identifying glioma drug candidates thereby informing new strategies for improving treatment.

bioinformatics↗

bioRxiv2024

Strategies for robust, accurate, and generalizable benchmarking of drug discovery platforms

Benchmarking is essential for the improvement and comparison of drug discovery platforms. We revised the protocols used to benchmark our Computational Analysis of Novel Drug Opportunities (CANDO) multiscale therapeutic discovery platform to bring them into strong alignment with best practices.CANDO ranked 7.4% and 12.1% of known drugs in the top 10 compounds for their respective diseases/indications using drug-indication mappings from the Comparative Toxicogenomics Database (CTD) and Therapeutic Targets Database (TTD), respectively. Better performance was weakly correlated (Spearman correlation coefficient >0.3) with the number of drugs associated with an indication and moderately correlated (coefficient >0.5) with intra-indication chemical similarity. There was also a moderate correlation between performance on our original and new benchmarking protocols. Higher performance was observed when using TTD instead of CTD when drug-indication associations appearing in both mappings were assessed. CANDO is available at https://github.com/ram-compbio/CANDO. The version used in this paper is available at http://compbio.buffalo.edu/data/mc_cando_benchmarking2. Supplementary data, drug-indication interaction matrices, and drug-indication mappings are available at http://compbio.buffalo.edu/data/mc_cando_benchmarking2.

bioinformatics↗

bioRxiv2022

Effective drug combinations targeting driver KRAS mutations in non-small cell lung cancer

1.Pharmacogenomics is a rapidly growing field with the goal of providing personalized care to every patient. Previously, we developed the Computational Analysis of Novel Drug Opportunities (CANDO) platform for multiscale therapeutic discovery to screen optimal compounds for any indication/disease by performing analytics on their interactions with large protein libraries. We implemented a comprehensive precision medicine drug discovery pipeline within the CANDO platform to determine which drugs are most likely to be effective against mutant phenotypes of non-small cell lung cancer (NSCLC) based on the supposition that drugs with similar interaction profiles (or signatures) will have similar behavior and therefore show synergistic effects. CANDO predicted that osimertinib, an EGFR inhibitor, is most likely to synergize with four KRAS inhibitors.Validation studies with cellular toxicity assays confirmed that osimertinib in combination with ARS-1620, a KRAS G12C inhibitor, and BAY-293, a pan-KRAS inhibitor, showed a synergistic effect on decreasing cellular proliferation by acting on mutant KRAS. Gene expression studies revealed that MAPK suppression is a key correlate of decreased cellular proliferation following treatment with KRAS inhibitor BAY-293, but not treatment with ARS-1620 or osimertinib. Our precision medicine pipeline may be used to identify compounds capable of synergizing with inhibitors of KRAS G12C, and to assess their likelihood of becoming drugs by understanding their behavior at the proteomic/interactomic scales.

bioinformatics↗

bioRxiv2022

Effective holistic characterization of small molecule effects using heterogeneous biological networks

The two most common reasons for attrition in therapeutic clinical trials are efficacy and safety. We integrated heterogeneous data to create a human interactome network that was used to comprehensively describe drug behavior in biological systems, with the goal of accurate therapeutic candidate generation. The Computational Analysis of Novel Drug Opportunities (CANDO) platform for shotgun multiscale therapeutic discovery, repurposing, and design was enhanced by integrating drug side effects, protein pathways, protein-protein interactions, protein-disease associations, and the Gene Ontology, complemented with its existing drug/compound, protein, and indication libraries. These integrated networks were reduced to a "multiscale interactomic signature" for each compound that describe its functional behavior as vectors of real values. These signatures are then used for relating compounds to each other with the hypothesis that similar signatures yield similar behavior. Our results indicated that there is significant biological information captured within our networks (particularly via side effects) which enhance the performance of our platform, as evaluated by performing all-against-all leave-one-out drug-indication association benchmarking. Further, drug impacts on pathways derived from computed compound-protein interaction scores served as the features for a random forest machine learning model trained to predict drug-indication associations, with applications to mental disorders and cancer metastasis highlighted. This interactomic pipeline highlights the ability of CANDO to accurately relate drugs in a multitarget and multiscale context, and paves the way for predicting novel putative drug candidates using the information gleaned from indirect data such as side effect profiles and protein pathway information.

bioinformatics↗

bioRxiv2022

Proteomic network analysis of bronchoalveolar lavage fluid in ex-smokers to discover implicated protein targets and novel drug treatments for chronic obstructive pulmonary disease

RationaleBronchoalveolar lavage of the epithelial lining fluid can sample the profound changes in the airway lumen milieu prevalent in Chronic Obstructive Pulmonary Disease (COPD). Characterizing the proteins in bronchoalveolar lavage fluid in COPD with advanced proteomic methods will identify disease-related changes, provide insight into pathogenetic mechanisms and potential therapeutics that will aid in the discovery of more effective therapeutics for COPD. ObjectivesWe compared epithelial lining fluid proteome of ex-smokers with moderate COPD who are not in exacerbation status COPD, to non-smoking healthy control subjects using advanced proteomics methods and applied proteome-scale translational bioinformatics approaches to identify potential therapeutic protein targets and drugs that modulate these proteins towards the treatment of COPD. MethodsProteomic profiles of bronchalveolar lavage fluid were obtained from 1) never-smoker control subjects with normal lung function (n=10) or 2) individuals with stable moderate (GOLD stage 2, FEV1 50% - 80% predicted) COPD who were ex-smokers for at least one year (n=10). NIHs Database for Annotation, Visualization and Integrated Discovery (DAVID) and Ingenuitys Ingenuity Pathway Analysis (IPA) were the two bioinformatics tools employed for network analysis on the differentially expressed proteins to identify potential crucial hub proteins. The drug-proteome interaction signature comparison and ranking approach implemented in the Computational Analysis of Novel Drug Opportunities (CANDO) platform for multiscale therapeutic discovery was utilized to identify potential repurposable drugs for the treatment of COPD based on the BALF proteome. Subsequently, a literature-based knowledge graph was utilized to rank combinations of drugs that would most likely ameloriate inflammatory processes by inhibition or activation of their functions. ResultsProteomic network analysis demonstrated that 233 of the >1800 proteins identified in the BALF were differentially expressed in COPD versus control, including proteins associated with inflammation, structural elements, and energy metabolism. Functional annotation of the differentially expressed proteins by their implicated biological processes, cellular localization, and transcription factor interactions was accomplished via DAVID. Canonical pathways containing the differential expressed proteins were detailed via the Ingenuity Pathway Analysis application. Topological network analysis demonstrated that four proteins act as central node proteins in the inflammatory pathways in COPD. The CANDO multiscale drug discovery platform was used to analyze the behavioral similarity between the interaction signatures of all FDA-approved drugs and the identified BALF proteins. The drugs with the signatures most similar interaction signatures to approved COPD drugs were extracted with the CANDO platform. The analysis revealed 189 drugs that putatively target the proteins implicated in COPD. The putative COPD drugs that were identified using CANDO were subsequently analyzed using a knowledge based technique to identify an optimal two drug combination that had the most appropriate effect on the central node proteins. ConclusionAnalysis of the BALF proteome revealed novel differentially expressed proteins in the epithelial lining fluid that elucidate COPD pathogenesis. Network analyses identified critical targets that have critical roles in modulating COPD pathogenesis, for which we identified several drugs that could be repurposed to treat COPD using a multiscale shotgun drug discovery approach.

bioinformatics↗

bioRxiv2021

Identifying Protein Features and Pathways Responsible for Toxicity using Machine learning, CANDO, and Tox21 datasets: Implications for Predictive Toxicology

Humans are exposed to numerous compounds daily, some of which have adverse effects on health. Computational approaches for modeling toxicological data in conjunction with machine learning algorithms have gained popularity over the last few years. Machine learning approaches have been used to predict toxicity-related biological activities using chemical structure descriptors. However, toxicity-related proteomic features have not been fully investigated. In this study, we construct a computational pipeline using machine learning models for predicting the most important protein features responsible for the toxicity of compounds taken from the Tox21 dataset that is implemented within the multiscale Computational Analysis of Novel Drug Opportunities (CANDO) therapeutic discovery platform. Tox21 is a highly imbalanced dataset consisting of twelve in vitro assays, seven from the nuclear receptor (NR) signaling pathway and five from the stress response (SR) pathway, for more than 10,000 compounds. For the machine learning model, we employed a random forest with the combination of Synthetic Minority Oversampling Technique (SMOTE) and Edited Nearest Neighbor (ENN) method (SMOTE+ENN), which is a resampling method to balance the activity class distribution. Within the NR and SR pathways, the activity of the aryl hydrocarbon receptor (NR-AhR), and mitochondrial membrane potential (SR-MMP) were two of the top-performing twelve toxicity endpoints with AUCROCs of 0.90 and 0.92, respectively. The top extracted features for evaluating compound toxicity were analyzed for enrichment to highlight the implicated biological pathways and proteins. We validated our enrichment results for the activity of the AhR using a thorough literature search. Our case study showed that the selected enriched pathways and proteins from our computational pipeline are not only correlated with AhR toxicity but also form a cascading upstream/downstream arrangement. Our work elucidate significant relationships between protein and compound interactions computed using CANDO and the associated biological pathways to which the proteins belong for twelve toxicity endpoints. This novel study uses machine learning not only to predict and understand toxicity but also elucidates therapeutic mechanisms at a proteomic level for a variety of toxicity endpoints.

bioinformatics↗

bioRxiv2021

A Deep Learning Proteomic Scale Approach for Drug Design

Computational approaches have accelerated novel therapeutic discovery in recent decades. The Computational Analysis of Novel Drug Opportunities (CANDO) platform for shotgun multi-target therapeutic discovery, repurposing, and design aims to improve their efficacy and safety by employing a holistic approach by computing interaction signatures between every drug/compound and a large library of non-redundant protein structures corresponding to the human proteome fold space. These signatures are compared and analyzed to determine if a given drug/compound is efficacious and safe for a given indication/disease. In this study, we used a deep learning based autoencoder to first reduce the dimensionality of CANDO computed drug-proteome interaction signatures. We then employed a reduced conditional variational autoencoder to generate novel drug-like compounds when given a target encoded "objective" signature. Using this model, we designed compounds to recreate the interaction signatures for twenty approved and experimental drugs and showed that 16/20 designed compounds are predicted to be significantly (p-value[≤] .05) more behaviorally similar relative to all corresponding controls, and 20/20 are predicted to be more behaviorally similar relative to a random control. We further observed that redesigns of objectives developed via rational drug design perform significantly better than those derived from natural sources (p-value[≤] .05), suggesting that the model has learned an abstraction of rational drug design. We also show that designed compounds are structurally diverse and synthetically feasible when compared to their respective objective drugs despite consistently high predicted behavioral similarity. Finally, we generated new designs that enhance thirteen drugs/compounds associated with non-small cell lung cancer and anti-aging properties using their predicted proteomic interaction signatures. This work represents a significant step forward in automating holistic therapeutic design with machine learning, and subsequently offers a reduction in the time needed to generate novel, effective, and safe drug leads for any indication.

bioinformatics↗

bioRxiv2020

Evaluating Performance of Drug Repurposing Technologies

Drug repurposing technologies are growing in number and maturing. However, comparison to each other and to reality is hindered due to lack of consensus with respect to performance evaluation. Such comparability is necessary to determine scientific merit and to ensure that only meaningful predictions from repurposing technologies carry through to further validation and eventual patient use. Here, we review and compare performance evaluation measures for these technologies using version 2 of our shotgun repurposing Computational Analysis of Novel Drug Opportunities (CANDO) platform to illustrate their benefits, drawbacks, and limitations. Understanding and using different performance evaluation metrics ensures robust cross platform comparability, enabling us to continuously strive towards optimal repurposing by decreasing time and cost of drug discovery and development.

bioinformatics↗

bioRxiv2019

cando.py: Open source software for analyzing large scale drug-protein-disease data

Elucidating drug-protein interactions is essential for understanding the beneficial effects of small molecule therapeutics in human disease states. Traditional drug discovery methods focus on optimizing the efficacy of a drug against a single biological target of interest. However, evidence supports the multitarget theory, i.e., drugs work by exerting their therapeutic effects via interaction with multiple biological targets. Analyzing drug interactions with a library of proteins provides further insight into disease systems while also allowing for prediction of putative therapeutics against specific indications. We present a Python package for analysis of drug-proteome and drug-disease relationships implementing the Computational Analysis of Novel Drug Opportunities (CANDO) platform [1-7]. The CANDO package allows for rapid drug similarity assessment, most notably via the bioinformatic docking protocol where billions of drug-protein interactions are rapidly scored and the similarity of drug-proteome interaction signatures is calculated. The package also implements a variety of bench-marking protocols to determine how well drugs are related to each other in terms of the indications/diseases for which they are approved. Drug predictions are generated through consensus scoring of the most similar compounds to drugs known to treat a particular indication. Support for comparing and ranking novel chemical entities, as well as machine learning modules for both benchmarking and putative drug candidate prediction is also available. The CANDO Python package is available on GitHub at https://github.com/ram-compbio/CANDO, through the Conda Python package installer, and at http://compbio.org/software/.

bioinformatics↗

bioRxiv2019

Exploration of interaction scoring criteria in the CANDO platform

BackgroundDrug discovery is an arduous process that requires many years and billions of dollars before approval for patient use. However, there are a number of drugs and human ingestibles approved for a variety of indications/diseases that can be potentially repurposed as new treatments for others, decreasing the time and cost required.\n\nMethodsCANDO (Computational Analysis of Novel Drug Opportunities) is a platform for shotgun, multitarget drug discovery and repurposing. The CANDO platform scores interactions between 46,784 proteins structures and 3,733 human use compounds using a bioinformatic docking protocol to generate compound-proteome interaction signatures that are then compared to identify candidates for repurposing. Benchmarking of the platform is accomplished by comparing the compound-proteome interaction signatures and determining whether signatures corresponding to pairs of drugs approved for the same indication fall within particular cutoffs.\n\nResultsWe have altered the scoring function of bioinformatic docking protocol in the newest version of our platform (v1.5) to use the best OBscore for each compound-protein interaction, resulting in an increased benchmarking accuracy from 11.7% in v1 to 12.8% in v1.5 for the top10 cutoff, the most stringent one used, and correspondingly from 24.9% to 31.2% for the top100 cutoff.\n\nConclusionsThe change in the interaction scoring and other bug fixes in CANDO v1.5 have resulted in improved benchmarking performance, making the platform more effective at predicting novel, therapeutic drug-indication pairs.

bioinformatics↗

bioRxiv2018

Identifying protein subsets and features responsible for improved drug repurposing accuracies using the CANDO platform

Drug repurposing is a valuable tool for combating the slowing rates of novel therapeutic discovery. The Computational Analysis of Novel Drug Opportunities (CANDO) platform performs shotgun repurposing of 3,733 drugs/compounds that map to 2,030 indications/diseases by predicting their interactions with 46,784 protein structures and relating them via proteomic interaction signatures. The accuracy of the CANDO platform is evaluated using our benchmarking protocol that assesses indication accuracies based on whether or not pairs of drugs associated with the same indication can be captured within a certain cutoff, which is a measure of the drug repurposing recovery rate. To identify subsets of proteins that exhibit the same therapeutic effectiveness as the full set, groups of 8 proteins were randomly selected and subsequently benchmarked 50 times. The resulting protein sets were ranked according to average indication accuracy, pairwise accuracy, and coverage (count of indications with non-zero accuracy). The best 50 subsets of 8 according to each metric were progressively combined into supersets after each iteration and benchmarked. These supersets yield up to 14% improvement in benchmarking accuracy, and represent a 100-1,000 fold reduction in the number of proteins relative to the full set. Protein supersets optimized using independent compound libraries derived from the full library were cross-tested and were shown to reproduce the performance relative to using all 46,784 proteins, indicating that these reduced size supersets are broadly applicable for characterizing drug behavior. Further analysis revealed that sets comprised of proteins with more equitably diverse ligand interactions are important for describing drug behavior. Our work elucidates the role of particular protein subsets and corresponding ligand interactions that play a role in computational drug repurposing, and paves the way for the use of machine learning approaches to further improve the accuracy of the CANDO platform and its repurposing potential.\n\nAuthor summaryDrug repurposing is a valuable approach for ameliorating the current problems plaguing drug discovery. We introduce a novel protein subset analysis pipeline that allows us to elucidate features important for drug repurposing accuracies using the Computational Analysis of Novel Drug Opportunities (CANDO) platform. Our platform relates drugs based on the similarity of their interactions with a diverse library of proteins. We subjected all proteins in the platform to a splitting and ranking protocol that ranked protein subsets based on their benchmarking performance. Further analysis of the best performing protein subsets revealed that the most useful proteins for describing how small molecule compounds behave in biological systems are those that are predicted to interact with a structurally diverse range of ligands. We hypothesize that this is a consequence of the multitarget nature of drugs and, conversely, the implied promiscuity of proteins in biological systems. These results may be used to make drug discovery more accurate and efficient by alleviating some of its bottlenecks, bringing us one step further in better understanding how drugs behave in the context of their environments.

bioinformatics↗