Analyzing Subject-Method Network of Bioinformatics and Biology

common and different research subjects and methods. Identifying popular and unpopular subjects and methods in both fields is critical to understand differences in two perspectives and to promote their collaboration. For Bioinformatics, discovering biologists' research interests that can be solved by bioinformatics is valuable. We explore subjects and methods dealt within Bioinformatics and Biology with an objective to understand the current state of the two research fields and to discover the potential hot spots for research. We introduce a subject-method network analysis and apply it to overview subjects, methods, and the associations between subjects and methods of the two fields.

For our analysis, we applied the categories of Web of Science Journal Citation Report(SJCR) that the data were obtained from Biology and Mathematical and Computational Biology categories for each Biology and Bioinformatics. We analyzed edges that a pair of "Mathematical modeling of biological systems" and "Theoretical ecology" ranked the third among edges. Ecologists were interested in interactions of all the creatures but had limitation in exploring and observing the whole system. However, recently, Bioinformatics enabled ecologists to build a biological systems as a mathematical model to simulate the whole system regarding environmental factors. The edge ranked at 8th of Bioinformatics is "genetic history" and "circadian rhythm" pair. It means scholars were interested in how climate affects genetic history.

We also analyzed communities that it showed difference in biology and bioinformatics. In the research of cancer, biology tries to understand the mechanism while bioinformatics searches the sequence to find the mutation that caused cancer. Both biology and bioinformatics semiparametric methods and meta-analysis, but biology applies statistical method and SVM to it while bioinformatics utilizes gene expression profiling, gene ontology, sequence alignment, protein prediction, computational methods in drug design, mass spectrometry frequently. It means biology looks cancer biology in molecular level but bioinformatics investigates it in gene level. It is similar to chronic disease.

This study is to build a higher level of knowledge structure called subject-method network to enhance our understanding on research problems and solutions in two relevant fields, Biology and Bioinformatics.

The extensions can be made by detecting subjects and methods with named entity recognition technique, by substituting degree centralities to other centralities, by adding meta-information such as author, affiliation, and journal, and by building an extensive network of subject-subject-method-method network.