[Neural network modeling of multitarget RAGE inhibitory activity]
| Autor: | D. T. Appazova, L. R. Yanaliyeva, Pavel M. Vassiliev, A. N. Kochetkov, Vladlen G. Klochkov, V. V. Vorfolomeyeva, Alexander A. Spasov |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
0301 basic medicine
endocrine system diseases Receptor for Advanced Glycation End Products Computational biology Inhibitory postsynaptic potential General Biochemistry Genetics and Molecular Biology 03 medical and health sciences 0302 clinical medicine Humans cardiovascular diseases Receptor Transcription factor Artificial neural network Kinase Chemistry NF-kappa B nutritional and metabolic diseases General Medicine Signal chain Molecular Docking Simulation 030104 developmental biology Docking (molecular) 030220 oncology & carcinogenesis cardiovascular system Neural Networks Computer Signal transduction human activities Signal Transduction |
| Zdroj: | Biomeditsinskaia khimiia. 65(2) |
| ISSN: | 2310-6972 |
| Popis: | Based on the methodology of artificial neural networks, models describing the dependence of the level of RAGE inhibitory activity on the affinity of compounds for target proteins of the RAGE-NF-kB signal pathway have been costructed. A validated database of the structures and activity levels of 183 known compounds, which were tested for RAGE inhibitory activity was formed. The analysis of the AGE-RAGE signaling pathways was carried out, 14 key RAGE-NF-kB signal pathway nodes were found, for which 34 relevant target proteins were identified. A database of 66 valid 3D models of 22 target proteins of the RAGE-NF-kB signal chain was compiled. Ensemble molecular docking of 3D models of 183 known RAGE inhibitors into sites of 66 valid 3D models of 22 relevant RAGE target proteins was performed and minimum docking energies for each compound were determined for each target. According to the method of artificial multilayer perceptron neural networks, classification models were constructed to predict level of RAGE inhibitory activity based on the calculated affinity of compounds for significant target proteins of the RAGE-NF-kB signaling chain. The prognostic ability of these models of RAGE-inhibitory activity was evaluated, the maximum accuracy according to ROC-analysis was 90% for a high level of activity. The sensitivity analysis of the developed multitarget models were carried out, the most significant targets of the RAGE-NF-kB signal transmission chain were determined. It was found that for high level of RAGE inhibitory activity, the most significant biotargets are not AGE receptors, but eight signaling kinases of the RAGE-NF-kB pathway and transcription factor NF-kB1. Thus, it is suggested that known compounds with high RAGE-inhibitory activity are preferential inhibitors of signal kinases.Na osnove metodologii iskusstvennykh neĭronnykh seteĭ postroeny modeli, opisyvaiushchie zavisimost' urovnia RAGE-ingibiruiushcheĭ aktivnosti ot affinnosti soedineniĭ k belkam-misheniam signal'nogo puti RAGE-NF-kB. Dlia étogo byla sformirovana validirovannaia baza dannykh po strukturam i urovniam aktivnosti 183 izvestnykh soedineniĭ, ispytannykh, po literaturnym dannym, na RAGE-ingibiruiushchuiu aktivnost'. Proveden analiz signal'nykh puteĭ AGE-RAGE, naĭdeny 14 osnovnykh uzlov signal'nogo puti RAGE-NF-kB, dlia kotorykh vyiavleny 34 relevantnykh belka-misheni. Sformirovana baza dannykh po 66 validnym 3D-modeliam 22 belkov-misheneĭ signal'noĭ tsepochki RAGE-NF-kB. Vypolnen ansamblevyĭ molekuliarnyĭ doking 183 izvestnykh ingibitorov RAGE v saĭty 66 validnykh 3D-modeleĭ 22 relevantnykh RAGE belkov-misheneĭ i opredeleny minimal'nye velichiny otsenochnoĭ funktsii dokinga dlia kazhdogo soedineniia v otnoshenii kazhdoĭ misheni. Po metodike iskusstvennykh mnogosloĭnykh pertseptronnykh neĭronnykh seteĭ postroeny klassifikatsionnye modeli dlia prognoza urovnia RAGE-ingibiruiushcheĭ aktivnosti po raschetnoĭ affinnosti soedineniĭ k znachimym belkam-misheniam signal'noĭ tsepi RAGE-NF-kB. Provedena otsenka prognosticheskoĭ sposobnosti étikh modeleĭ; maksimal'naia tochnost' po dannym ROC analiza sostavila 90% dlia vysokogo urovnia aktivnosti. Vypolnen analiz chuvstvitel'nosti razrabotannykh neĭrosetevykh mul'titargetnykh modeleĭ, opredeleny naibolee znachimye misheni tsepochki peredachi signala RAGE-NF-kB. Naĭdeno, chto dlia vysokogo urovnia RAGE-ingibiruiushcheĭ aktivnosti naibolee znachimymi biomisheniami iavliaiutsia ne AGE-retseptory, a vosem' signal'nykh kinaz puti RAGE-NF-kB i transkriptsionnyĭ faktor NF-kB1. Predpolozheno, chto izvestnye soedineniia s vysokoĭ RAGE-ingibiruiushcheĭ aktivnost'iu po suti iavliaiutsia preimushchestvenno ingibitorami signal'nykh kinaz. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|