Dual-Target Drug Discovery in Alzheimer\u2019s Disease Using Machine Learning guided approach integrated with structure based virtual screening<\/strong><\/td><\/tr>Asim Ayesha1,2<\/sup>, Jastrz\u0119bski Micha\u0142 K.1,2<\/sup>, Karcz Tadeusz2<\/sup>, Olejarz-Maciej Agnieszka2<\/sup>, Koval Maryna1<\/sup>, Kuku\u0142a-Koch Wirginia3<\/sup>, A.Kaczor Agnieszka4<\/sup><\/em>
1<\/sup><\/em> Doctoral School of the Medical University of Lublin, Poland;
2<\/sup><\/em> Faculty of Pharmacy, Jagiellonian University, Medical College, Cracow, Poland;
3<\/sup><\/em> Department of Pharmacognosy with Medical Plants Garden, Medical University of Lublin, Poland;
4<\/sup><\/em> Department of Synthesis and Chemical Technology of Pharmaceutical Substances with Computer Modeling Laboratory, Medical University of Lublin, Poland<\/td><\/tr>Abstract<\/strong>
Alzheimer\u2019s disease (AD) is a progressive neurodegenerative disorder with complex and multifactorial pathology, for which current therapies provide only symptomatic relief. Dual inhibition of two key targets, acetylcholinesterase (AChE) and monoamine oxidase-B (MAO-B), has emerged as a promising strategy to concurrently enhance cholinergic neurotransmission and reduce the oxidative stress in AD [1]. In the present study, a machine learning\u2013guided approach integrated with structure-based virtual screening was employed to identify novel small-molecule candidates as potential dual AChE and MAO-B inhibitors.
Bioactivity data for both targets was retrieved from the ChEMBL database and, after curation and preprocessing, used to develop predictive machine learning models employing artificial neural networks (ANNs) and decision trees (DTs). These models were applied to virtually screen the Enamine database of approximately 0.46 million compounds, yielding 3,890 prioritized candidates. This subset was subsequently subjected to high-throughput virtual screening against X-ray structures of AChE and MAO-B. As a result, compounds showing favorable binding affinities and key interactions with critical active-site residues of both enzymes were shortlisted, yielding 11 potential dual-target candidates, alongside 18 compounds prioritized as selective MAO-B inhibitors. All the compounds were further evaluated using PAINS filtering and pharmacokinetic profiling. Experimental validation showed that compound 5 demonstrated balanced dual inhibition of AChE and MAO-B, supporting its potential as a promising lead scaffold. Additionally, selective inhibition was observed, with compound 2 exhibiting 32% AChE inhibition (berberine: 57%), and compound 10 showing 96% MAO-B inhibition comparable to safinamide (100%). Moreover, two compounds from selective MAO-B inhibitors series exhibited strong MAO-B inhibition (96\u201398%), approaching the activity of clinically used inhibitors such as rasagiline and safinamide. Collectively, these results validate the effectiveness of the integrated machine learning and structure-based virtual screening workflow and identify promising chemical scaffolds for the rational development of dual AChE\/MAO-B inhibitors. In vivo studies, including zebrafish assays and behavioral testing in mice are currently underway to further validate the therapeutic potential of these candidates. This work was supported by the National Science Center (NCN, Poland) under the OPUS grant 2021\/43\/B\/NZ7\/01732.<\/td><\/tr>References <\/strong>
[1] Asim, A.; Jastrz\u0119bski, M.K.; Kaczor, A.A. Dual inhibitors of acetylcholinesterase and monoamine oxidase-B for the treatment of Alzheimer\u2019s disease. Molecules 2025, 30, 2975. https:\/\/doi.org\/10.3390\/molecules30142975.
[2] Asim, A.; Jastrz\u0119bski, M.K.; Kaczor, A.A. Dual inhibitors of acetylcholinesterase and monoamine oxidase-B for the treatment of Alzheimer\u2019s disease. Molecules 2025, 30, 2975. https:\/\/doi.org\/10.3390\/molecules30142975.<\/td><\/tr><\/tbody><\/table><\/div><\/figure>\n\n\n\n<\/p>\n","protected":false},"excerpt":{"rendered":" OC9 – Ayesha Asim Medical University of Lublin Poland e-mail Dual-Target Drug Discovery in Alzheimer\u2019s Disease Using Machine Learning guided approach integrated with structure based virtual screening Asim Ayesha1,2, Jastrz\u0119bski Micha\u0142 K.1,2, Karcz Tadeusz2, Olejarz-Maciej Agnieszka2, […]<\/p>\n","protected":false},"author":9643,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[],"class_list":["post-1662","post","type-post","status-publish","format-standard","hentry","category-senza-categoria"],"_links":{"self":[{"href":"https:\/\/sites.unica.it\/pe2026\/wp-json\/wp\/v2\/posts\/1662","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/sites.unica.it\/pe2026\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/sites.unica.it\/pe2026\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/sites.unica.it\/pe2026\/wp-json\/wp\/v2\/users\/9643"}],"replies":[{"embeddable":true,"href":"https:\/\/sites.unica.it\/pe2026\/wp-json\/wp\/v2\/comments?post=1662"}],"version-history":[{"count":6,"href":"https:\/\/sites.unica.it\/pe2026\/wp-json\/wp\/v2\/posts\/1662\/revisions"}],"predecessor-version":[{"id":1832,"href":"https:\/\/sites.unica.it\/pe2026\/wp-json\/wp\/v2\/posts\/1662\/revisions\/1832"}],"wp:attachment":[{"href":"https:\/\/sites.unica.it\/pe2026\/wp-json\/wp\/v2\/media?parent=1662"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/sites.unica.it\/pe2026\/wp-json\/wp\/v2\/categories?post=1662"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/sites.unica.it\/pe2026\/wp-json\/wp\/v2\/tags?post=1662"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}} | | | |
![\"\"](\"http:\/\/sites.unica.it\/pe2026\/files\/2025\/09\/cropped-00_Profile_Tondo-Social__Spilla_PE26_1200px.png\")
<\/figure>