Artificial Intelligence in Cancer Research and Precision Medicine

Cancer Discovery, Journal Year: 2021, Volume and Issue: 11(4), P. 900 - 915

Published: April 1, 2021

Artificial intelligence (AI) is rapidly reshaping cancer research and personalized clinical care. Availability of high-dimensionality datasets coupled with advances in high-performance computing, as well innovative deep learning architectures, has led to an explosion AI use various aspects oncology research. These applications range from detection classification cancer, molecular characterization tumors their microenvironment, drug discovery repurposing, predicting treatment outcomes for patients. As these start penetrating the clinic, we foresee a shifting paradigm care becoming strongly driven by AI. SIGNIFICANCE: potential dramatically affect nearly all oncology-from enhancing diagnosis personalizing discovering novel anticancer drugs. Here, review recent enormous progress application oncology, highlight limitations pitfalls, chart path adoption clinic.

Language: Английский

---

SynergyFinder Plus: Toward Better Interpretation and Annotation of Drug Combination Screening Datasets

Genomics Proteomics & Bioinformatics, Journal Year: 2022, Volume and Issue: 20(3), P. 587 - 596

Published: Jan. 25, 2022

Abstract Combinatorial therapies have been recently proposed to improve the efficacy of anticancer treatment. The SynergyFinder R package is a software used analyze pre-clinical drug combination datasets. Here, we report major updates for improved interpretation and annotation screening results. Unlike existing implementations, updated includes five main innovations. 1) We extend mathematical models higher-order data analysis implement dimension reduction techniques visualizing synergy landscape. 2) provide statistical sensitivity with confidence intervals P values. 3) incorporate barometer harmonize multiple scoring methods consensus metric synergy. 4) evaluate an unbiased clinical potential. 5) enable fast drugs cell lines, including their chemical target information. These annotations will mechanisms action combinations. To facilitate use within discovery community, also web server at www.synergyfinderplus.org as user-friendly interface more flexible versatile data.

Language: Английский

---

A review on machine learning approaches and trends in drug discovery

Computational and Structural Biotechnology Journal, Journal Year: 2021, Volume and Issue: 19, P. 4538 - 4558

Published: Jan. 1, 2021

Drug discovery aims at finding new compounds with specific chemical properties for the treatment of diseases. In last years, approach used in this search presents an important component computer science skyrocketing machine learning techniques due to its democratization. With objectives set by Precision Medicine initiative and challenges generated, it is necessary establish robust, standard reproducible computational methodologies achieve set. Currently, predictive models based on Machine Learning have gained great importance step prior preclinical studies. This stage manages drastically reduce costs research times drugs. review article focuses how these are being recent years research. Analyzing state art field will give us idea where cheminformatics be developed short term, limitations positive results has achieved. focus mainly methods model molecular data, as well biological problems addressed algorithms drug years.

Language: Английский

---

Effective drug combinations in breast, colon and pancreatic cancer cells

Nature, Journal Year: 2022, Volume and Issue: 603(7899), P. 166 - 173

Published: Feb. 23, 2022

Abstract Combinations of anti-cancer drugs can overcome resistance and provide new treatments 1,2 . The number possible drug combinations vastly exceeds what could be tested clinically. Efforts to systematically identify active the tissues molecular contexts in which they are most effective accelerate development combination treatments. Here we evaluate potency efficacy 2,025 clinically relevant two-drug combinations, generating a dataset encompassing 125 molecularly characterized breast, colorectal pancreatic cancer cell lines. We show that synergy between is rare highly context-dependent, targeted agents likely synergistic. incorporate multi-omic features biomarkers specify synergistic their contexts, including basal-like breast cancer, microsatellite-stable or KRAS -mutant colon cancer. Our results irinotecan CHEK1 inhibition have effects – TP53 double-mutant cells, leading apoptosis suppression tumour xenograft growth. This study identifies distinct subpopulations resource guide rational efforts develop combinatorial

Language: Английский

---

Artificial intelligence in drug discovery: what is realistic, what are illusions? Part 1: Ways to make an impact, and why we are not there yet

Drug Discovery Today, Journal Year: 2020, Volume and Issue: 26(2), P. 511 - 524

Published: Dec. 18, 2020

Although artificial intelligence (AI) has had a profound impact on areas such as image recognition, comparable advances in drug discovery are rare. This article quantifies the stages of which improvements time taken, success rate or affordability will have most overall bringing new drugs to market. Changes clinical rates improving discovery; other words, quality decisions regarding compound take forward (and how conduct trials) more important than speed cost. current AI focus make given compound, question make, using efficacy and safety-related end points, received significantly less attention. As consequence, proxy measures available data cannot fully utilize potential discovery, particular when it comes safety vivo. Thus, addressing questions generate points model be key clinically relevant decision-making future.

Language: Английский

---

Independent Drug Action in Combination Therapy: Implications for Precision Oncology

Cancer Discovery, Journal Year: 2022, Volume and Issue: 12(3), P. 606 - 624

Published: Jan. 4, 2022

Combination therapies are superior to monotherapy for many cancers. This advantage was historically ascribed the ability of combinations address tumor heterogeneity, but synergistic interaction is now a common explanation as well design criterion new combinations. We review evidence that independent drug action, described in 1961, explains efficacy practice-changing combination therapies: it provides populations patients with heterogeneous sensitivities multiple chances benefit from at least one drug. Understanding response heterogeneity could reveal predictive or pharmacodynamic biomarkers more precise use existing drugs and realize benefits additivity synergy.Significance:. The model action represents an effective means predict magnitude likely be observed clinical trials therapies. "bet-hedging" strategy implicit suggests individual often only subset—sometimes one—of combination. Personalized, targeted therapy, consisting agents active particular patient, will increase, perhaps substantially, therapeutic benefit. Precision approaches this type require better understanding variability biomarkers, which entail preclinical research on diverse panels cancer models rather than studying synergy unusually sensitive models.

Language: Английский

---

DrugCombDB: a comprehensive database of drug combinations toward the discovery of combinatorial therapy

Nucleic Acids Research, Journal Year: 2019, Volume and Issue: unknown

Published: Oct. 17, 2019

Drug combinations have demonstrated high efficacy and low adverse side effects compared to single drug administration in cancer therapies thus drawn intensive attention from researchers pharmaceutical enterprises. Due the rapid development of high-throughput screening (HTS), number combination datasets available has increased tremendously recent years. Therefore, there is an urgent need for a comprehensive database that crucial both experimental computational synergistic combinations. In this paper, we present DrugCombDB, devoted curation various data sources: (i) HTS assays combinations; (ii) manual curations literature; (iii) FDA Orange Book external databases. Specifically, DrugCombDB includes 448 555 derived assays, covering 2887 unique drugs 124 human cell lines. particular, more than 6000 000 quantitative dose responses which computed multiple synergy scores determine overall or antagonistic addition extracted existing databases, manually curated 457 thousands PubMed publications. To benefit further validation models, are ready train prediction models classification regression analysis were constructed other significant related gathered. A website with user-friendly graphical visualization been developed users access wealth download prebuilt datasets. Our at http://drugcombdb.denglab.org/.

Language: Английский

---

Deep learning for drug response prediction in cancer

Briefings in Bioinformatics, Journal Year: 2019, Volume and Issue: 22(1), P. 360 - 379

Published: Dec. 17, 2019

Abstract Predicting the sensitivity of tumors to specific anti-cancer treatments is a challenge paramount importance for precision medicine. Machine learning(ML) algorithms can be trained on high-throughput screening data develop models that are able predict response cancer cell lines and patients novel drugs or drug combinations. Deep learning (DL) refers distinct class ML have achieved top-level performance in variety fields, including discovery. These types unique characteristics may make them more suitable complex task modeling based both biological chemical data, but application DL prediction has been unexplored until very recently. The few studies published shown promising results, use beginning attract greater interest from researchers field. In this article, we critically review recently employed methods lines. We also provide brief description main architectures used these studies. Additionally, present selection publicly available resources models. Finally, address limitations approaches discussion possible paths further improvement. Contact: [email protected]

Language: Английский

---

Big data in basic and translational cancer research

Nature reviews. Cancer, Journal Year: 2022, Volume and Issue: 22(11), P. 625 - 639

Published: Sept. 5, 2022

Language: Английский

---

Deep learning identifies synergistic drug combinations for treating COVID-19

Proceedings of the National Academy of Sciences, Journal Year: 2021, Volume and Issue: 118(39)

Published: Sept. 15, 2021

Significance COVID-19 has caused more than 2.5 million deaths worldwide. It is imperative that we develop therapies can mitigate the effect of disease. While searching for individual drugs this purpose been met with difficulties, synergistic drug combinations offer a promising alternative. However, lack high-quality training data pertaining to makes it challenging use existing machine learning methods effective novel combination prediction tasks. Our proposed approach addresses challenge by leveraging additional readily available data, such as drug−target interactions, thus enabling an in silico search against SARS-CoV-2.

Language: Английский

---