Object
Title: Heuristic and statistical QSAR and QSPR models of compounds with serotonergic activity
Abstract:
Serotonin is considered an important component of the regulation of many processes in the human body, affecting mood, appetite, libido, memory, attention, cognitive function, pain transmission, platelet aggregation and peristalsis, along with the secretion of digestive juices in the gastrointestinal tract. Serotonin receptors and the serotonin transporter are a point of entry for many drugs used in neurological disorders and also for drugs with antiemetic effects. Due to their limited efficacy, as well as the occurrence of frequent adverse effects, it is important to search for new structures that could become new drugs in the future. Drug discovery is a time consuming and resource intensive process. The rise of Artificial Intelligence is helping to enhance its efficiency and effectiveness. One important aspect is the creation of Quantitative Structure-Activity Relationship (QSAR) models. In order for a therapeutic substance to interact with a selected binding site the affinity of the molecule to the biological target is a paramount factor. Therefore, QSAR models are useful tools for improving the process of finding high-affinity molecules. From the literature review, it is apparent that there is a deficiency of QSAR models for serotonin receptors and transporter based on a large dataset extensively covering the structural diversity of the molecules. The purpose of this study w ; as to create QSAR models for eleven serotonin receptors (5-HT1A, 5-HT1B, 5-HT1D, 5-HT2A, 5-HT2B, 5-HT2C, 5-HT3, 5-HT4, 5-HT5A, 5-HT6 and 5-HT7) and the serotonin transporter. Once the QSAR models were created, it was important to implement them into an easy-to-use web application so that researchers around the world could conveniently use the resulting models. The application, called SerotoninAI, runs on computers and mobile devices using the most popular web browsers. The tool has been developed to also incorporate modules related to permeability across the intestinal barrier, which provides information for oral administration, and crossing the blood-brain barrier, which is important for substances intended to affect the central nervous system. The last two modules are related to serotonergic activity, making it possible to predict whether a molecule might exhibit such activity against serotonin receptors, and suggesting selective binding to one of eleven serotonin receptor subtypes. The models were created based on two-dimensional Mordred descriptors using the Automated Machine Learning tool Mljar. For each of the created models, SHAP (SHapley Additive exPlanations) analysis was conducted indicating the most important descriptors and their direction of influence on affinity, permeability, serotonergic activity and selectivity towards serotonin receptors.
Place of publishing:
Level of degree:
Degree discipline:
Degree grantor:
Rada Dyscypliny Nauki farmaceutyczne
Promoter:
Mendyk, Aleksander ; Pacławski, Adam
Date issued:
Identifier:
Call number:
Language:
Access rights:
Object collections:
Last modified:
Dec 13, 2024
In our library since:
Dec 13, 2024
Number of object content hits:
0
Number of object content views in PDF format
0
All available object's versions:
http://dl.cm-uj.krakow.pl:8080/publication/5175
Show description in RDF format:
Show description in OAI-PMH format:
| Edition name | Date |
|---|---|
| ZB-141075 | Dec 13, 2024 |
Objects
Similar
Łapińska, Natalia
Żmudzki, Paweł
Kalicińska, Jadwiga
Sudoł-Tałaj, Sylwia
Kaczorowska, Katarzyna
Kołaczkowski, Marcin
