Abstract
In this paper, we propose a framework, based on Multi-Fractal De-trended Fluctuation Cross-Correlation Analysis (MF-DCCA), to investigate the cross-correlation features between the molecular properties and aqueous solubility using the data set of Delaney-processed widely used for Deep Learning (DL). The molecular properties of Minimum Degree, Number of H-Bond Donors, Number of Rings, Number of Rotatable bonds and Polar Surface Area have a weak power-law cross-correlation with aqueous solubility; the Molecular Weight has a weak long-range cross-correlation. Obvious oscillation of long-range cross-correlation captures through a sliding window approach between molecular properties and aqueous solubility based on the pseudo-time series. Experimental comparison of time-varying cross-correlation widths shows that window widths are more robust to time-varying Hurst exponent. This work can provide a reference for reducing the data dimension for neural network molecular property prediction tasks and help in high-accuracy predicting molecular aqueous solubility on DL.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: Physica A: Statistical Mechanics and its Applications
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
