Last modified: 2015.VII.19.14:08 - MIAÚ-RSS
A MIAÚ immár 20­+ éve áll a Köz szolgálatára!

The evaluation of exogenous ligands cross-reactivity to a 7TM receptor based on online artificial intelligence engines

Leading article: 2020. January (MIAU No. 257.)
(Previous article: MIAU No. 256.)

Keywords: similarity, robot-biologist, robot-chemist, Turing-test

Abstract: For every organism from a cell to human it is a general concept that, to survive you have to adapt to the changing environment. To fulfil this requirement, receiving signals coming from the surroundings which are going to generate a response is needed. So, the basic model of signalling consists of a producer, a signal, a target and a transduction pathway which is going to be activated when the special signal has been identified. In cellular signalling processes the acceptor is a molecule called a receptor. In this study we worked with motilin receptor (MLNR) which is a membrane integrated protein and the natural ligand of it is a peptide hormone called motilin (MLN). This MLN-MLNR system has a pivotal function in smooth muscle contraction and the exact mechanisms in its signalling pathway is relatively well described. MLNR is the member of the 7TM receptor superfamily. There is a huge interest in 7TMs since they are coded by around 1000 genes in mammalian genomes and around 30% of FDA approved drugs act on them. In life science research it is often necessary to evaluate whether there are other molecules - beside its natural ligand(s) - which can interact with a given receptor. These compounds can be endogens as well as exogens. Regarding animal welfare and financial considerations, the only possible way to execute such a huge screen is in silico modelling. There are several methods for this topic including virtual screening, machine learning, QSAR analysis etc. Performing these techniques are often a challenging task since they are very computational and time consuming. Our goal was to try to establish a relatively easy-to-perform-method which can facilitate to explore novel compounds for a given target. This goal could be approximated based on similarity analyses, where the arbitrary detailed descriptions (attributes) of the potential objects (molecules) could be interpreted in an aggregated way. The optimized analyses proved whether each potential molecule can be seen as the same concerning the complex (context free) similarity index. The multi-layered validated result is atilmotin – which means this molecule can be evaluated as the most similar compared to the molecule of motilin – where the validation based on symmetry of staircase functions like in case of artificial intelligence-oriented term generation processes. More (DOC) *** More (PDF)

Please, send Your comments per email!

miau.my-x.hu

myxfree.tool

rss.services