Active Search for Computer‐aided Drug Design

Oglic, Dino; Oatley, Steven A.; Macdonald, Simon J. F.; McInally, Thomas; Garnett, Roman; Hirst, Jonathan D.; Gärtner, Thomas

doi:10.1002/minf.201700130

Cited by 21 publications

(25 citation statements)

References 65 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…, x T } to maximize the number of targets identified, y t . Many real-world problems can be naturally posed in terms of active search; drug discovery [4,12,13], materials discovery [6], and product recommendation [15] are a few examples.…”

Section: Introductionmentioning

confidence: 99%

Efficient nonmyopic active search with applications in drug and materials discovery

Jiang¹,

Malkomes²,

Moseley³

et al. 2018

Preprint

Self Cite

View full text Add to dashboard Cite

Active search is a learning paradigm for actively identifying as many members of a given class as possible. A critical target scenario is high-throughput screening for scientific discovery, such as drug or materials discovery. In this paper 1 , we approach this problem in Bayesian decision framework. We first derive the Bayesian optimal policy under a natural utility, and establish a theoretical hardness of active search, proving that the optimal policy can not be approximated for any constant ratio. We also study the batch setting for the first time, where a batch of b > 1 points can be queried at each iteration. We give an asymptotic lower bound, linear in batch size, on the adaptivity gap: how much we could lose if we query b points at a time for t iterations, instead of one point at a time for bt iterations. We then introduce a novel approach to nonmyopic approximations of the optimal policy that admits efficient computation. Our proposed policy can automatically trade off exploration and exploitation, without relying on any tuning parameters. We also generalize our policy to batch setting, and propose two approaches to tackle the combinatorial search challenge. We evaluate our proposed policies on a large database of drug discovery and materials science. Results demonstrate the superior performance of our proposed policy in both sequential and batch setting; the nonmyopic behavior is also illustrated in various aspects.1 This paper summarizes the contributions of two papers published at ICML 2017[6] and accepted at NIPS 2018 [7]. Proofs, related work, and some experimental results are omitted.32nd Conference on Neural Information Processing Systems (NIPS 2018),

show abstract

Section: Introductionmentioning

confidence: 99%

Efficient nonmyopic active search with applications in drug and materials discovery

Jiang¹,

Malkomes²,

Moseley³

et al. 2018

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Computer science has been identified as a useful approach to explore drug effectiveness and new compounds with specific features in computer‐assisted drug discovery and design, which greatly increases the efficiency of drug design (Oglic et al ., 2018). With the explosive growth of drug data, drug prediction tools such as docking (Barradas‐Bautista et al ., 2018) and machine learning (de Avila et al ., 2017) provide a theoretical framework for the discovery and prioritization of bioactive compounds with desired pharmacological effects and the optimization of those compounds as drug‐like leads.…”

Section: Discussionmentioning

confidence: 99%

Hispaglabridin B, a constituent of liquorice identified by a bioinformatics and machine learning approach, relieves protein‐energy wasting by inhibiting forkhead box O1

Huang

Wang

et al. 2018

British J Pharmacology

View full text Add to dashboard Cite

Background and Purpose Liquorice is the root of Glycyrrhiza glabra, which is a popular food in Europe and China that has previously shown benefits for skeletal fatigue and nutrient metabolism. However, the mechanism and active ingredients remain largely unclear. The aim of this study was to investigate the active ingredients of liquorice for muscle wasting and elucidate the underlying mechanisms. Experimental Approach RNA‐Seq and bioinformatics analysis were applied to predict the main target of liquorice. A machine learning model and a docking tool were used to predict active ingredients. Isotope labelling experiments, immunostaining, Western blots, qRT‐PCR, ChIP‐PCR and luciferase reporters were utilized to test the pharmacological effects in vitro and in vivo. The reverse effects were verified through recombination‐based overexpression. Key Results The liposoluble constituents of liquorice improved muscle wasting by inhibiting protein catabolism and fibre atrophy. We further identified FoxO1 as the target of liposoluble constituents of liquorice. In addition, hispaglabridin B (HB) was predicted as an inhibitor of FoxO1. Further studies determined that HB improved muscle wasting by inhibiting catabolism in vivo and in vitro. HB also markedly suppressed the transcriptional activity of FoxO1, with decreased expression of the muscle‐specific E3 ubiquitin ligases MuRF1 and Atrogin‐1. Conclusions and Implications HB can serve as a novel natural food extract for preventing muscle wasting in chronic kidney disease and possibly other catabolic conditions.

show abstract

“…One approach used to increase the effectiveness in the development of new drugs is the use of computer-aided drug design (CADD, well known as an in silico method) techniques, which uses a computational chemistry approach for the drug discovery process. CADD is a cyclic process for developing new drugs, in which all stages of design and analysis are performed by computer programs, operated by medicinal chemists (Oglic et al, 2018).…”

Section: Computer-aided Drug Design (Cadd)mentioning

confidence: 99%