本文選自《藥學(xué)進(jìn)展》2021年第7期,作者劉曉凡 1,孫翔宇 1,朱迅 2* ?!端帉W(xué)進(jìn)展》雜志是由中國藥科大學(xué)和中國藥學(xué)會共同主辦、國家教育部主管
(1. 火石創(chuàng)造,浙江 杭州 310051;2. 吉林大學(xué)基礎(chǔ)醫(yī)學(xué)院,吉林 長春 130021)
與 AI 在其他場景的應(yīng)用類似,AI+ 新藥研發(fā)的實現(xiàn)路徑包括五大流程:1)獲取目標(biāo)訓(xùn)練數(shù)據(jù)集;2)AI 自主學(xué)習(xí)算法建模;3)多次訓(xùn)練優(yōu)化模型;4)測試集應(yīng)用以評估模型性能;5)基于模型實現(xiàn)分子篩選、預(yù)測、分析等預(yù)定目標(biāo)。算法、數(shù)據(jù)集和模型這 3 個要素是必不可少的部分,其中,算法和數(shù)據(jù)是實現(xiàn)應(yīng)用的關(guān)鍵。
圖1:人工智能+新藥研發(fā)企業(yè)圖譜
來源:火石創(chuàng)造根據(jù)公開資料整理
來源:火石創(chuàng)造根據(jù)公開資料整理
參考資料:
Terry M. IQVIA report: biopharma R&D investment has increased32% over last 5 years[EB/OL].(2019-04-23)[2021-06-24].https://www.biospace.com/article/report-biopharma-focuses-on-oncologyrare-diseases-and-orphan-drugs/.
Fleming N. How artificial intelligence is changing drugdiscovery[EB/OL].(2018-05-10)[2021-06-05]. https://www.nature.com/articles/d41586-018-05267-x.
Taylor P. Pharma is 'getting lower returns on R&D’[EB/OL].(2017-12-15)[2021-05-21]. http://www.pmlive.com/pharma_news/pharma_is_getting_lower_returns_on_r_and_d_1214594.
劉伯炎 , 王群 , 徐俐穎 , 等 . 人工智能技術(shù)在醫(yī)藥研發(fā)中的應(yīng)用[J]. 中國新藥雜志 , 2020, 29(14): 1979–1986.
? Wong C H, Siah K W, Lo A W. Estimation of clinical trial successrates and related parameters[J]. Biostatistics, 2019, 20(2): 273–286.
陸濤 . 人工智能在醫(yī)藥領(lǐng)域的應(yīng)用 [J]. 藥學(xué)進(jìn)展 , 2019, 44(1):1–3.
梁禮 , 鄧成龍 , 張艷敏 , 等 . 人工智能在藥物發(fā)現(xiàn)中的應(yīng)用與挑戰(zhàn) [J]. 藥學(xué)進(jìn)展 , 2020, 44(1): 18–27.
Mcculloch W S, Pitts W. A logical calculus of the ideas immanent innervous activity[J]. Bull Math Biophys, 1943, 5(4): 115–133.
Lu H M, Li Y J, Uemura T, et al. FDCNet: filtering deepconvolutional network for marine organism classification[J].Multimed Tools Appl, 2017, 77(17): 21847–21860.
Yang X, Wang Y F, Byrne R, et al. Concepts of artificial intelligencefor computer-assisted drug discovery[J]. Chem Rev, 2019, 119(18):10520–10594.
Chen H M, Engkvist O, Wang Y H, et al. The rise of deep learningin drug discovery[J]. Drug Discov Today, 2018, 23(6): 1241–1250.
Lavecchia A. Deep learning in drug discovery: opportunities,challenges and future prospects[J]. Drug Discov Today, 2019,24(10): 2017–2032.
張建楠 , 李瑩瑩 , 張燁琳 . AI 驅(qū)動新藥研發(fā)深度發(fā)展 [EB/OL].(2021-01-27)[2021-06-28]. http://www.imit.org.cn/index.php?g=&m=article&a=index&id=356&cid=11.
搜狐網(wǎng) . 正大天晴與阿里云合作,AI 制藥提高化合物篩選準(zhǔn)確率 20% 新藥研發(fā)困難重重?正大天晴與阿里云合作用 AI 尋找解決辦法 [EB/OL]. (2018-09-04)[2021-06-30]. https://www.sohu.com/a/251826274_114877.
Willems H, De Cesco S, Svensson F. Computational chemistryon a budget: supporting drug discovery with limited resources:miniperspective[J].J Med Chem, 2020, 63(18): 10158–10169.
Chan H, Shan H, Dahoun T, et al. Advancing drug discovery viaartificial intelligence[J]. Trends Pharmacol Sci, 2019, 40(8): 592–604.
Zhavoronkov A. Artificial intelligence for drug discovery, biomarkerdevelopment, and generation of novel chemistry[J]. Mol Pharm,2018, 15(10): 4311–4313.
Hong R S, Mattei A, Sheikh A Y, et al. Novel physics-basedensemble modeling approach that utilizes 3D molecular conformationand packing to access aqueous thermodynamic solubility: a casestudy of orally available bromodomain and extraterminal domaininhibitor lead optimization series[J]. J Chem Inf Model, 2021, 61(3):1412–1426.
Lee P K. Shortcut strategy for screening compounds with clinicalpotentials for drug development[EB/OL].(2018-11-03)[2021-06-26].https://www.eurekalert.org/pub_releases/2018-12/cuoh-ssf120318.php.
Lin X, Duan X, Jacobs C, et al. High-throughput brain activitymapping and machine learning as a foundation for systemsneuropharmacology[J]. Nat Commun, 2018, 9(1): 5142. DOI:10.1038/s41573-019-0024-5.
Vamathevan J, Clark D, Czodrowski P, et al. Applications ofmachine learning in drug discovery and development[J]. Nat RevDrug Discov, 2019(18): 463–477.
Bender A, Cortés-Ciriano I. 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[J]. Drug Discov Today, 2021,26(2):511–524.
Xiao Q, Wang L, Supekar S, et al. Structure of human steroid5α-reductase 2 with the anti-androgen drug finasteride[J]. NatCommun, 2020, 11(1): 5430. DOI:10.1038/s41467-020-19249-z.
Liu R, Wei L, Zhang P. A deep learning framework for drugrepurposing via emulating clinical trials on real-world patientdata[J]. Nat Mach Intell, 2021(3): 68–75.
Ulenaers J. The impact of artificial intelligence on the right to a fairtrial: towards a robot judge?[J]. Asian J Law Econ, 2020, 11(2):20200008. DOI:10.1515/ajle-2020-0008.
朱壽華 , 凌澤農(nóng) , 周金花 . 人工智能技術(shù)在醫(yī)療健康領(lǐng)域的應(yīng)用 [J]. 電子技術(shù)與軟件工程 , 2020, 171(1): 18–19.
央廣網(wǎng) . 中國人工智能人才缺口超 500 萬供求比例僅為 1:10[EB/OL]. (2017-07-14)[2021-06-30]. https://baijiahao.baidu.com/s?id=1572854410914751&wfr=spider&for=pc.
中國日報網(wǎng) . 法律該賦予人工智能什么地位 ? 實現(xiàn)與人的智能疊 拼 [EB/OL]. (2018-01-23)[2021-06-30]. https://baijiahao.baidu.com/s?id=1590352506664610651&wfr=spider&for=pc.
Bhatt A. Artificial intelligence in managing clinical trial design andconduct: man and machine still on the learning curve?[J]. PerspectClin Res, 2021, 12(1): 1. DOI:10.4103/picr.PICR_312_20.
Jiménez-Luna J, Grisoni F, Weskamp N, et al. Artificial intelligencein drug discovery: recent advances and future perspectives[EB/OL].(2021-04-02)[2021-06-28]. http://doi-org-443.group10-s.ccame.net/10.1080/17460441.2021.1909567.
聯(lián)系客服