Picture of blood cells

Open Access research which pushes advances in bionanotechnology

Strathprints makes available scholarly Open Access content by researchers in the Strathclyde Institute of Pharmacy & Biomedical Sciences (SIPBS) , based within the Faculty of Science.

SIPBS is a major research centre in Scotland focusing on 'new medicines', 'better medicines' and 'better use of medicines'. This includes the exploration of nanoparticles and nanomedicines within the wider research agenda of bionanotechnology, in which the tools of nanotechnology are applied to solve biological problems. At SIPBS multidisciplinary approaches are also pursued to improve bioscience understanding of novel therapeutic targets with the aim of developing therapeutic interventions and the investigation, development and manufacture of drug substances and products.

Explore the Open Access research of SIPBS. Or explore all of Strathclyde's Open Access research...

3D matters! 3D-RISM and 3D convolutional neural network for accurate bioaccumulation prediction

Sosnin, Sergey and Misin, Maksim and Palmer, David S. and Fedorov, Maxim V. (2018) 3D matters! 3D-RISM and 3D convolutional neural network for accurate bioaccumulation prediction. Journal of Physics: Condensed Matter, 30. ISSN 0953-8984

[img] Text (Sosnin-etal-JPCM-2018-3D-RISM-and-3D-convolutional-neural-network-for-accurate-bioaccumulation-prediction)
Accepted Author Manuscript
Restricted to Repository staff only until 19 July 2019.

Download (3MB) | Request a copy from the Strathclyde author


    In this work, we present a new method for predicting complex physicalchemical properties of organic molecules. The approach utilizes 3D convolutional neural network (ActivNet4) that uses solvent spatial distributions around solutes as input. These spatial distributions are obtained by a molecular theory called threedimensional reference interaction site model (3D-RISM). We have shown that the method allows one to achieve a good accuracy of prediction of bioconcentration factor (BCF) which is difficult to predict by direct application of methods of molecular theory or simulations. Our research demonstrates that combination of molecular theories with modern machine learning approaches can be effectively used for predicting properties that are otherwise inaccessible to purely theory-based models.