Kvantno-kemijsko dokiranje heterocikla u aktivno mjesto butirilkolinesteraze primjenom dubokog pojačanog učenja

master's thesis

• Title (english): Quantum-chemical docking of heterocycles into the active site of butyrylcholinesterase using deep reinforcement learning
• Author: Nika Jakobović
• Mentor: Tomica Hrenar (mentor)
• Committee member: Tomica Hrenar (predsjednik povjerenstva)
• Committee member: Aleksandra Maršavelski (član povjerenstva)
• Committee member: Ivan Kodrin (član povjerenstva)
• Committee member: Branimir Bertoša (član povjerenstva)
• Granter: University of Zagreb; Faculty of Science; (Department of Chemistry); Zagreb
• Defense date and country: 2024-07-19, Croatia
• Scientific / art field, discipline and subdiscipline: NATURAL SCIENCES; Chemistry

Abstract

Alzheimerova bolest teška je i neizlječiva neurološka bolest koju karakterizira gubitak kognitivnih sposobnosti pojedinca. Simptomi uključuju demenciju te promjene u ponašanju i osobnosti bolesnika, a bolest predstavlja jedan od vodećih uzroka smrti u svijetu. Prema kolinergičkoj hipotezi, u mozgu oboljelih osoba smanjena je koncentracija neurotransmitera acetilkolina te se za povećanje njegove koncentracije kao lijekovi koriste inhibitori acetilkolinesteraze i butirilkolinesteraze. Ti inhibitori u svojoj strukturi često sadrže heterocikličke sustave te su u ovom radu istraženi konfiguracijski prostori malih heterocikličkih molekula u aktivnom mjestu butirilkolinesteraze. Molekule aziridina i azetidina te njihove kombinacije dokirane su korištenjem kvantno-kemijskog Monte Carlo dokiranja. Za pretraživanje konfiguracijskih prostora primijenjen je protokol računalstva s velikom propusnošću, a za 100 struktura najniže energije provedene su dodatne optimizacije geometrije i izračuni standardnih Gibbsovih energija vezanja. Pokazano je da se energija dokiranih molekula u aktivnom mjestu butirilkolinesteraze može uspješno procijeniti korištenjem duboke neuronske mreže istrenirane algoritmom dubokoga pojačanog učenja.

Abstract (english)

Alzheimer's disease is a severe and incurable neurological disease that is characterized by the loss of an individual's cognitive abilities. Symptoms are dementia and changes in the patient's behavior and personality, and the disease is one of the leading causes of death in the world. According to the cholinergic hypothesis, the concentration of the neurotransmitter acetylcholine is reduced in the brain of affected individuals, and acetylcholinesterase and butyrylcholinesterase inhibitors are used as drugs to increase its concentration. These inhibitors often contain heterocyclic systems in their structure, and in this work the configurational spaces of small heterocyclic molecules in the active site of butyrylcholinesterase were investigated. Aziridine and azetidine molecules and their combinations were docked using quantum- chemical Monte Carlo docking. A high-throughput computing protocol was applied to search the configurational spaces, and additional geometry optimizations as well as calculation of standard Gibbs binding energies were performed for the 100 lowest energy structures. It has been shown that the energy of docked molecules in the active site of butyrylcholinesterase can be successfully estimated using a deep neural network trained by deep reinforcement learning.

• Keywords: Alzheimerova bolest; butirilkolinesteraza; kvantno-kemijsko dokiranje; duboko pojačano učenje
• Keywords (english): Alzheimer's disease; butyrylcholinesterase; quantum-chemical docking; deep reinforcement learning
• Language: croatian
• URN:NBN: urn:nbn:hr:217:630197
• Project: Number: IP-2022-10-9525; Title: Metom usmjerena sinteza inhibitora kolinesteraza podržana strojnim učenjem; Title: Target-guided synthesis of cholinesterase inhibitors supported by machine learning; Acronym: TargetCh; Leader: Ines Primožič; Jurisdiction: Croatia; Funder: Hrvatska zaklada za znanost; Funding stream: Research Projects
• Study programme: Title: Graduate university studies in chemistry; specializations in: Graduate university studies in chemistry; Course: Graduate university studies in chemistry; Study programme type: university; Study level: graduate; Academic / professional title: sveučilišni/a magistar/magistra kemije (sveučilišni/a magistar/magistra kemije)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Embargoed access; Embargo expiration date: 2025-07-19
• Created on: 2024-10-02 13:43:30