Graph kernels for chemoinformatics – a critical discussion

Graph kernels for chemoinformatics – a critical discussion

We analyze the use, advantages, and drawbacks of graph kernels in chemoin-formatics, including a comparison of kernel-based approaches with other methodology, as well as examples of applications. Kernel-based machine learning [1], now widely applied in chemoinformatics, delivers state-of-the-art performance [2] in tasks like classification and regression. Molecular graph kernels [3] are a recen...

Graph Kernels for Chemoinformatics

In chemoinformatics and bioinformatics, it is effective to automatically predict the properties of chemical compounds and proteins with computeraided methods, since this can substantially reduce the costs of research and development by screening out unlikely compounds and proteins from the candidates for ‘wet” experiment. Data-driven predictive modeling is one of the main research topics in che...

Graph kernels in chemoinformatics

Two New Graph Kernels and Applications to Chemoinformatics

Chemoinformatics is a well established research field concerned with the discovery of molecule’s properties through informational techniques. Computer science’s research fields mainly concerned by the chemoinformatics field are machine learning and graph theory. From this point of view, graph kernels provide a nice framework combining machine learning techniques with graph theory. Such kernels ...

Graph Mining in Chemoinformatics

In the first step of drug discovery process, a large number of lead compounds are found by high throughput screening. To identify physicochemical properties of the lead compounds, SAR and QSAR analyses are commonly applied (Gasteiger & Engel, 2003). In machine learning terminology, SAR is understood as a classification task where a chemical compound is given as an input, and the learning machin...