We are perhaps at a turning point for making cheminformatics accessible to scientists who are not computational chemists. The proliferation of mobile devices has seen the development of software or ‘apps' that can be used for sophisticated chemistry workflows. These apps can offer capabilities to the practicing chemist that are approaching those of conventional desktop-based software, whereby each app focuses on a relatively small range of tasks. Mobile apps that can pull in and integrate public content from many sources relating to molecules and data are also being developed. Apps for drug discovery are already evolving rapidly and are able to communicate with each other to create composite workflows of increasing complexity, enabling informatics aspects of drug discovery (i. e. accessing data, modeling and visualization) to be done anywhere by potentially anyone. We will describe how these cheminformatics apps can be used productively and some of the future opportunities that we envision.

View full abstract

One of the most critical and frequent problems in biomedical data classification is imbalanced class distribution, where samples from the majority class significantly outnumber the minority class. SMOTE is a well-known general over-sampling method used to address this problem; however, in some cases it cannot improve or even reduces classification performance. To address these issues, we have developed a novel minority over-sampling method named safe-SMOTE. Experimental results from two gene expression datasets for cancer classification (i.e., colon-cancer and leukemia) and six imbalanced benchmark datasets from the UCI Machine Learning Repository showed that our method achieved better sensitivity and G-mean values than both the control method (i.e., no over-sampling) and SMOTE. For example, in the colon-cancer dataset, although the sensitivity and specificity achieved by SMOTE (81.36% and 88.63%) were lower than for the control method (81.59% and 89.50%), safe-SMOTE in contrast had these values increase (81.82% and 90.50%). Similarly, the G-mean value of the control (85.45%) decreased to 84.91% when SMOTE was employed, but increased to 86.04% when using safe-SMOTE. In the leukemia dataset, SMOTE was able to improve the sensitivity and G-mean values with respect to the control; however, safe-SMOTE achieved noticeable, even greater improvements for both of these criteria.

View full abstract

Four human neuraminidases (hNEUs1-4) have been identified. Among them, hNEU1 has been studied extensively as a target for sialidosis. It has been desired to understand the biological functions of hNEU1 at the molecular and atomic levels. The three-dimensional structure of hNEU1 is not known at present. In the present work, we constructed a three-dimensional structure of hNEU1 by homology modeling, and then performed correlation analyses between observed and calculated free-energy changes (quantitative structure−activity relationship (QSAR) analyses), coupled with LERE (linear expression by representative energy terms) procedure using the modeled three-dimensional structure in order to confirm the validity of the modeled structure. The atomic coordinates of all atoms in the verified model of hNEU1 are available. The proposed structure of hNEU1 will be useful and helpful for further studies concerning the biological and chemical functions of hNEU1. The present article is one of continuous works derived from the one that won the CBI Award for the best presentation in the CBI/JSBi 2011 Joint Conference (presented by Seiji Hitaoka).

View full abstract

The ab initio fragment molecular orbital-based molecular dynamics (FMO-MD) method was extended for simulation of solvated polypeptides by the introduction of an algorithm named dynamic fragmentation with static fragments (DF/SF). In FMO-MD, the force acting on each nucleus is calculated by the FMO method, which requires fragmentation of the simulated molecule. The fragmentation data must be redefined, depending on the time-dependent change of the molecular configuration, and the DF/SF algorithm governs this redefinition. In the DF/SF algorithm, some fragments are manually classified as static and unchanged, while others are considered dynamic and subject to change. Various options of the algorithm were implemented in the ABINIT-MP program. The options were tested and discussed as they applied to FMO-MD simulations of the solvated (Gly)₂ dipeptide, in which the two amino acid residues of the peptide were regarded as static (invariable) while surrounding water molecules were regarded as dynamic (variable). Future prospects for the FMO-MD simulation of biopolymers are discussed based upon the tests of the DF/SF algorithm.

View full abstract