Harnessing machine learning to find synergistic combinations for FDA-approved cancer drugs

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Jan. 29, 2024

Abstract Combination therapy is a fundamental strategy in cancer chemotherapy. It involves administering two or more anti-cancer agents to increase efficacy and overcome multidrug resistance compared monotherapy. However, drug combinations can exhibit synergy, additivity, antagonism. This study presents machine learning framework classify predict combinations. The utilizes several key steps including data collection annotation from the O’Neil interaction dataset, preprocessing, stratified splitting into training test sets, construction evaluation of classification models categorize as synergistic, additive, antagonistic, application regression combination sensitivity scores for enhanced predictions prior work, last step examination features mechanisms action understand synergy behaviors optimal identified pairs most likely synergize against different cancers. Kinase inhibitors combined with mTOR inhibitors, DNA damage-inducing drugs HDAC showed benefit, particularly ovarian, melanoma, prostate, lung colorectal carcinomas. Analysis highlighted Gemcitabine, MK-8776 AZD1775 frequently synergizing across types. provides valuable approach uncover effective multi-drug regimens.

Language: Английский

---

Potentials and future perspectives of multi-target drugs in cancer treatment: the next generation anti-cancer agents

Cell Communication and Signaling, Journal Year: 2024, Volume and Issue: 22(1)

Published: April 15, 2024

Abstract Cancer is a major public health problem worldwide with more than an estimated 19.3 million new cases in 2020. The occurrence rises dramatically age, and the overall risk accumulation combined tendency for cellular repair mechanisms to be less effective older individuals. Conventional cancer treatments, such as radiotherapy, surgery, chemotherapy, have been used decades combat cancer. However, emergence of novel fields research has led exploration innovative treatment approaches focused on immunotherapy, epigenetic therapy, targeted multi-omics, also multi-target therapy. hypothesis was based that drugs designed act against individual targets cannot usually battle multigenic diseases like Multi-target therapies, either combination or sequential order, recommended acquired intrinsic resistance anti-cancer treatments. Several studies multi-targeting treatments due their advantages include; overcoming clonal heterogeneity, lower multi-drug (MDR), decreased drug toxicity, thereby side effects. In this study, we'll discuss about drugs, benefits improving recent advances field multi-targeted drugs. Also, we will study performed clinical trials using therapeutic agents treatment.

Language: Английский

---

The role and potential of computer-aided drug discovery strategies in the discovery of novel antimicrobials

Computers in Biology and Medicine, Journal Year: 2024, Volume and Issue: 169, P. 107927 - 107927

Published: Jan. 2, 2024

Antimicrobial resistance (AMR) has become more of a concern in recent decades, particularly infections associated with global public health threats. The development new antibiotics is crucial to ensuring infection control and eradicating AMR. Although drug discovery are essential processes the transformation candidate from laboratory bedside, they often very complicated, expensive, time-consuming. pharmaceutical sector continuously innovating strategies reduce research costs accelerate candidates. Computer-aided (CADD) emerged as powerful promising technology that renews hope researchers for faster identification, design, cheaper, less resource-intensive, efficient In this review, we discuss an overview AMR, potential, limitations CADD AMR discovery, case studies successful application technique rapid identification various This review will aid achieving better understanding available techniques novel candidates against resistant pathogens other infectious agents.

Language: Английский

---

Predicting drug combination response surfaces

npj Drug Discovery., Journal Year: 2025, Volume and Issue: 2(1)

Published: Feb. 3, 2025

Abstract Prediction of drug combination responses is a research question growing importance for cancer and other complex diseases. Current machine learning approaches generally consider predicting either synergy summaries or single dose-response values, which fail to appropriately model the continuous nature underlying surface can lead inconsistencies when score matrix reconstructed from separate predictions. We propose novel prediction method, comboKR, that directly predicts response combination. The method based on powerful input–output kernel regression technique functional modelling surface. ComboKR belongs family output methods, where target function, in our case, non-linear parametric Our thus avoids discretized forms as scalars, vectors matrices, therefore provides better interpolation extrapolation along surfaces. As an important part approach, we develop normalisation between surfaces standardises heterogeneous experimental designs used measure dose-responses, allows training with data measured different laboratories. experiments two predictive scenarios using datasets highlight suitability proposed approach especially traditionally challenging setting new drugs not available data.

Language: Английский

---

Artificial intelligence-driven drug development against autoimmune diseases

Trends in Pharmacological Sciences, Journal Year: 2023, Volume and Issue: 44(7), P. 411 - 424

Published: May 31, 2023

Artificial intelligence (AI)-based predictive models are being used to foster a precision medicine approach treat complex chronic diseases such as autoimmune and autoinflammatory disorders (AIIDs). In the past few years first of systemic lupus erythematosus (SLE), primary Sjögren syndrome (pSS), rheumatoid arthritis (RA) have been produced by molecular profiling patients using omic technologies integrating data with AI. These advances confirmed pathophysiology involving multiple proinflammatory pathways also provide evidence for shared dysregulation across different AIIDs. I discuss how stratify patients, assess causality in pathophysiology, design drug candidates silico, predict efficacy virtual patients. By relating individual patient characteristics predicted properties millions candidates, these can improve management AIIDs through more personalized treatments.

Language: Английский

---

Single-cell transcriptomes identify patient-tailored therapies for selective co-inhibition of cancer clones

Nature Communications, Journal Year: 2024, Volume and Issue: 15(1)

Published: Oct. 3, 2024

Language: Английский

---

Emerging Promise of Computational Techniques in Anti-Cancer Research: At a Glance

Bioengineering, Journal Year: 2022, Volume and Issue: 9(8), P. 335 - 335

Published: July 25, 2022

Research on the immune system and cancer has led to development of new medicines that enable former attack cells. Drugs specifically target destroy cells are horizon; there also drugs use specific signals stop multiplying. Machine learning algorithms can significantly support increase rate research complicated diseases help find remedies. One area medical study could greatly benefit from machine is exploration genomes discovery best treatment protocols for different subtypes disease. However, developing a drug time-consuming, complicated, dangerous, costly. Traditional production take up 15 years, costing over USD 1 billion. Therefore, computer-aided design (CADD) emerged as powerful promising technology develop quicker, cheaper, more efficient designs. Many technologies methods have been introduced enhance productivity analytical methodologies, they become crucial part many programs; scanning programs, example, ligand screening structural virtual techniques hit detection optimization. In this review, we examined various types computational focusing anticancer drugs. Machine-based in basic translational reach levels personalized medicine marked by speedy advanced data analysis still beyond reach. Ending know it means ensuring every patient access safe effective therapies. Recent developments had large remarkable impact yielded useful insights into field therapy. With an emphasis medications, covered components paper. Transcriptomics, toxicogenomics, functional genomics, biological networks only few examples bioinformatics used forecast medications combinations based multi-omics data. We believe general review databases now available today will be beneficial creation approaches.

Language: Английский

---

A Review on the Recent Applications of Deep Learning in Predictive Drug Toxicological Studies

Chemical Research in Toxicology, Journal Year: 2023, Volume and Issue: 36(8), P. 1174 - 1205

Published: Aug. 10, 2023

Drug toxicity prediction is an important step in ensuring patient safety during drug design studies. While traditional preclinical studies have historically relied on animal models to evaluate toxicity, recent advances deep-learning approaches shown great promise advancing science and reducing use However, deep-learning-based also face challenges handling large biological data sets, model interpretability, regulatory acceptance. In this review, we provide overview of developments for predicting highlighting their potential advantages over methods the need address limitations. Deep-learning demonstrated excellent performance outcomes from various sources such as chemical structures, genomic data, high-throughput screening assays. The deep learning automated feature engineering discussed. This review emphasizes ethical concerns related studies, including reduction Furthermore, emerging applications prediction, drug–drug interactions rare subpopulations, are highlighted. integration with discussed a way develop more reliable efficient predictive assessment, paving safer effective discovery development. Overall, highlights critical role toxicology evaluation, emphasizing continued research development rapidly evolving field. By addressing limitations methods, leveraging engineering, concerns, revolutionize improve

Language: Английский

---

DrugMAP 2.0: molecular atlas and pharma-information of all drugs

Nucleic Acids Research, Journal Year: 2024, Volume and Issue: 53(D1), P. D1372 - D1382

Published: Sept. 13, 2024

Abstract The escalating costs and high failure rates have decelerated the pace of drug development, which amplifies research interests in developing combinatorial/repurposed drugs understanding off-target adverse reaction (ADR). In other words, it is demanded to delineate molecular atlas pharma-information for interactions. However, such invaluable data were inadequately covered by existing databases. this study, a major update was thus conducted DrugMAP, accumulated (a) 20831 combinatorial their interacting involving 1583 pharmacologically important molecules; (b) 842 repurposed with 795 (c) 3260 off-targets relevant ADRs 2731 (d) various types pharmaceutical information, including diverse ADMET properties, versatile diseases, ADRs/off-targets. With growing demands discovering therapies rapidly emerging interest AI-based discovery, DrugMAP highly expected act as an indispensable supplement databases facilitating accessible at: https://idrblab.org/drugmap/.

Language: Английский

---

The recent progress of deep-learning-based in silico prediction of drug combination

Drug Discovery Today, Journal Year: 2023, Volume and Issue: 28(7), P. 103625 - 103625

Published: May 25, 2023

Language: Английский

---