1
|
Xu X, Zhou J, Zhu C, Zhan Q, Li Z, Zhang R, Wang Y, Liao X, Gao X. Optimization of binding affinities in chemical space with generative pre-trained transformer and deep reinforcement learning. F1000Res 2024; 12:757. [PMID: 38434657 PMCID: PMC10905145 DOI: 10.12688/f1000research.130936.2] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Accepted: 02/15/2024] [Indexed: 03/05/2024] Open
Abstract
Background The key challenge in drug discovery is to discover novel compounds with desirable properties. Among the properties, binding affinity to a target is one of the prerequisites and usually evaluated by molecular docking or quantitative structure activity relationship (QSAR) models. Methods In this study, we developed SGPT-RL, which uses a generative pre-trained transformer (GPT) as the policy network of the reinforcement learning (RL) agent to optimize the binding affinity to a target. SGPT-RL was evaluated on the Moses distribution learning benchmark and two goal-directed generation tasks, with Dopamine Receptor D2 (DRD2) and Angiotensin-Converting Enzyme 2 (ACE2) as the targets. Both QSAR model and molecular docking were implemented as the optimization goals in the tasks. The popular Reinvent method was used as the baseline for comparison. Results The results on the Moses benchmark showed that SGPT-RL learned good property distributions and generated molecules with high validity and novelty. On the two goal-directed generation tasks, both SGPT-RL and Reinvent were able to generate valid molecules with improved target scores. The SGPT-RL method achieved better results than Reinvent on the ACE2 task, where molecular docking was used as the optimization goal. Further analysis shows that SGPT-RL learned conserved scaffold patterns during exploration. Conclusions The superior performance of SGPT-RL in the ACE2 task indicates that it can be applied to the virtual screening process where molecular docking is widely used as the criteria. Besides, the scaffold patterns learned by SGPT-RL during the exploration process can assist chemists to better design and discover novel lead candidates.
Collapse
Affiliation(s)
- Xiaopeng Xu
- Computational Bioscience Research Center (CBRC), King Abdullah University of Science and Technology (KAUST), Thuwal, 23955-6900, Saudi Arabia
- Computer, Electrical and Mathematical Sciences and Engineering (CEMSE), King Abdullah University of Science and Technology (KAUST), Thuwal, 23955-6900, Saudi Arabia
| | - Juexiao Zhou
- Computational Bioscience Research Center (CBRC), King Abdullah University of Science and Technology (KAUST), Thuwal, 23955-6900, Saudi Arabia
- Computer, Electrical and Mathematical Sciences and Engineering (CEMSE), King Abdullah University of Science and Technology (KAUST), Thuwal, 23955-6900, Saudi Arabia
| | - Chen Zhu
- KAUST Catalysis Center (KCC), King Abdullah University of Science and Technology (KAUST), Thuwal, 23955-6900, Saudi Arabia
| | - Qing Zhan
- Computational Bioscience Research Center (CBRC), King Abdullah University of Science and Technology (KAUST), Thuwal, 23955-6900, Saudi Arabia
- Computer, Electrical and Mathematical Sciences and Engineering (CEMSE), King Abdullah University of Science and Technology (KAUST), Thuwal, 23955-6900, Saudi Arabia
| | - Zhongxiao Li
- Computational Bioscience Research Center (CBRC), King Abdullah University of Science and Technology (KAUST), Thuwal, 23955-6900, Saudi Arabia
- Computer, Electrical and Mathematical Sciences and Engineering (CEMSE), King Abdullah University of Science and Technology (KAUST), Thuwal, 23955-6900, Saudi Arabia
| | | | - Yu Wang
- Syneron Technology, Guangzhou, China
| | - Xingyu Liao
- Computational Bioscience Research Center (CBRC), King Abdullah University of Science and Technology (KAUST), Thuwal, 23955-6900, Saudi Arabia
- Computer, Electrical and Mathematical Sciences and Engineering (CEMSE), King Abdullah University of Science and Technology (KAUST), Thuwal, 23955-6900, Saudi Arabia
| | - Xin Gao
- Computational Bioscience Research Center (CBRC), King Abdullah University of Science and Technology (KAUST), Thuwal, 23955-6900, Saudi Arabia
- Computer, Electrical and Mathematical Sciences and Engineering (CEMSE), King Abdullah University of Science and Technology (KAUST), Thuwal, 23955-6900, Saudi Arabia
| |
Collapse
|
2
|
Gaa R, Kumari K, Mayer HM, Yanakieva D, Tsai SP, Joshi S, Guenther R, Doerner A. An integrated mammalian library approach for optimization and enhanced microfluidics-assisted antibody hit discovery. Artif Cells Nanomed Biotechnol 2023; 51:74-82. [PMID: 36762883 DOI: 10.1080/21691401.2023.2173219] [Citation(s) in RCA: 2] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 02/11/2023]
Abstract
Recent years have seen the development of a variety of mammalian library approaches for display and secretion mode. Advantages include library approaches for engineering, preservation of precious immune repertoires and their repeated interrogation, as well as screening in final therapeutic format and host. Mammalian display approaches for antibody optimization exploit these advantages, necessitating the generation of large libraries but in turn enabling early screening for both manufacturability and target specificity. For suitable libraries, high antibody integration rates and resulting monoclonality need to be balanced - we present a solution for sufficient transmutability and acceptable monoclonality by applying an optimized ratio of coding to non-coding lentivirus. The recent advent of microfluidic-assisted hit discovery represents a perfect match to mammalian libraries in secretion mode, as the lower throughput fits well with the facile generation of libraries comprising a few million functional clones. In the presented work, Chinese Hamster Ovary cells were engineered to both express the target of interest and secrete antibodies in relevant formats, and specific clones were strongly enriched by high throughput screening for autocrine cellular binding. The powerful combination of mammalian secretion libraries and microfluidics-assisted hit discovery could reduce attrition rates and increase the probability to identify the best possible therapeutic antibody hits faster.
Collapse
Affiliation(s)
- Ramona Gaa
- Protein Engineering and Antibody Technologies, Merck KGaA, Darmstadt, Germany
| | - Kavita Kumari
- Discovery Biology, Syngene International, Phase-IV, Bangalore, India
| | - Hannah Melina Mayer
- Protein Engineering and Antibody Technologies, Merck KGaA, Darmstadt, Germany
| | - Desislava Yanakieva
- Protein Engineering and Antibody Technologies, Merck KGaA, Darmstadt, Germany
| | - Shang-Pu Tsai
- Protein Engineering and Antibody Technologies, EMD Serono, Billerica, MA, USA
| | - Saurabh Joshi
- Discovery Biology, Syngene International, Phase-IV, Bangalore, India
| | - Ralf Guenther
- Protein Engineering and Antibody Technologies, Merck KGaA, Darmstadt, Germany
| | - Achim Doerner
- Protein Engineering and Antibody Technologies, Merck KGaA, Darmstadt, Germany
| |
Collapse
|
3
|
Moreira-Filho JT, Neves BJ, Cajas RA, Moraes JD, Andrade CH. Artificial intelligence-guided approach for efficient virtual screening of hits against Schistosoma mansoni. Future Med Chem 2023; 15:2033-2050. [PMID: 37937522 DOI: 10.4155/fmc-2023-0152] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/09/2023] Open
Abstract
Background: The impact of schistosomiasis, which affects over 230 million people, emphasizes the urgency of developing new antischistosomal drugs. Artificial intelligence is vital in accelerating the drug discovery process. Methodology & results: We developed classification and regression machine learning models to predict the schistosomicidal activity of compounds not experimentally tested. The prioritized compounds were tested on schistosomula and adult stages of Schistosoma mansoni. Four compounds demonstrated significant activity against schistosomula, with 50% effective concentration values ranging from 9.8 to 32.5 μM, while exhibiting no toxicity in animal and human cell lines. Conclusion: These findings represent a significant step forward in the discovery of antischistosomal drugs. Further optimization of these active compounds can pave the way for their progression into preclinical studies.
Collapse
Affiliation(s)
- José Teófilo Moreira-Filho
- Laboratory of Molecular Modeling and Drug Design (LabMol), Faculdade de Farmácia, Universidade Federal de Goiás, Goiânia, 74605-170, Brazil
| | - Bruno Junior Neves
- Laboratory of Molecular Modeling and Drug Design (LabMol), Faculdade de Farmácia, Universidade Federal de Goiás, Goiânia, 74605-170, Brazil
| | - Rayssa Araujo Cajas
- Research Center on Neglected Diseases (NPDN), Universidade Guarulhos, Guarulhos, 07023-070, Brazil
| | - Josué de Moraes
- Research Center on Neglected Diseases (NPDN), Universidade Guarulhos, Guarulhos, 07023-070, Brazil
| | - Carolina Horta Andrade
- Laboratory of Molecular Modeling and Drug Design (LabMol), Faculdade de Farmácia, Universidade Federal de Goiás, Goiânia, 74605-170, Brazil
- Center for the Research and Advancement in Fragments and molecular Targets (CRAFT), School of Pharmaceutical Sciences at Ribeirao Preto, University of São Paulo, Ribeirão Preto, SP, Brazil
| |
Collapse
|
4
|
Abstract
INTRODUCTION After the WHO declared Zika virus (ZIKV) as a public health emergency of international concern, intense research for the development of vaccines and drugs has been undertaken, leading to the development of several candidates. Areas covered: This review discusses the developments achieved so far by computational methods in the discovery of candidate compounds targeting ZIKV proteins, i.e. the envelope and capsid structural proteins, the NS3 helicase/protease, and the NS5 methyltransferase/RNA-dependent RNA polymerase. Expert opinion: Research for effective drugs against ZIKV is still in a very early discovery phase. Notwithstanding the intense efforts for the development of new drugs and the identification of several promising candidates by using different approaches, including computational methods, so far only a few candidates have been experimentally tested. An important caveat of anti-flavivirus drug development is represented by the difficult of reproducing the in vivo microenvironment of the replication complex, which may lead to discrepancies between in vitro results and experimental evaluation in vivo. Moreover, anti-ZIKV drugs have the additional requirement of an excellent safety profile in pregnancy and ability to diffuse to different tissues, including the central nervous system, the testis, and the placenta.
Collapse
Affiliation(s)
| | - Silvia Riccetti
- a Department of Molecular Medicine , University of Padova , Padova , Italy
| | - Marta Trevisan
- a Department of Molecular Medicine , University of Padova , Padova , Italy
| | - Luisa Barzon
- a Department of Molecular Medicine , University of Padova , Padova , Italy
| |
Collapse
|
5
|
Saucedo-Mendiola ML, Salas-Pacheco JM, Nájera H, Rojo-Domínguez A, Yépez-Mulia L, Avitia-Domínguez C, Téllez-Valencia A. Discovery of Entamoeba histolytica hexokinase 1 inhibitors through homology modeling and virtual screening. J Enzyme Inhib Med Chem 2013; 29:325-32. [PMID: 23534932 DOI: 10.3109/14756366.2013.779265] [Citation(s) in RCA: 10] [Impact Index Per Article: 0.9] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/13/2022] Open
Abstract
Entamoeba histolytica, the parasite which causes amebiasis is responsible for 110,000 deaths a year. Entamoeba histolytica depends on glycolysis to obtain ATP for cellular work. According to metabolic flux studies, hexokinase exerts the highest flux control of this metabolic pathway; therefore, it is an excellent target in the search of new antiamebic drugs. To this end, a tridimensional model of E. histolytica hexokinase 1 (EhHK1) was constructed and validated by homology modeling. After virtual screening of 14,400 small molecules, the 100 with the best docking scores were selected, purchased and assessed in their inhibitory capacity. The results showed that three molecules (compounds 2921, 11275 and 2755) inhibited EhHK1 with an I50 of 48, 91 and 96 µM, respectively. Thus, we found the first inhibitors of EhHK1 that can be used in the search of new chemotherapeutic agents against amebiasis.
Collapse
Affiliation(s)
- María Leticia Saucedo-Mendiola
- Facultad de Ciencias Químicas, Universidad Juárez del Estado de Durango, Av. Veterinaria S/N Circuito Universitario , Durango , México
| | | | | | | | | | | | | |
Collapse
|