A Novel Algorithm for the Precise Calculation of the Maximal Information Coefficient

Measuring associations is an important scientific task. A novel measurement method maximal information coefficient (MIC) was proposed to identify a broad class of associations. As foreseen by its authors, MIC implementation algorithm ApproxMaxMI is not always convergent to real MIC values. An algori...

Descripción completa

Detalles Bibliográficos
Autores principales: Zhang, Yi, Jia, Shili, Huang, Haiyun, Qiu, Jiqing, Zhou, Changjie
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Nature Publishing Group 2014
Materias:
Article
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4200418/
https://www.ncbi.nlm.nih.gov/pubmed/25322794
http://dx.doi.org/10.1038/srep06662
Descripción
Sumario:Measuring associations is an important scientific task. A novel measurement method maximal information coefficient (MIC) was proposed to identify a broad class of associations. As foreseen by its authors, MIC implementation algorithm ApproxMaxMI is not always convergent to real MIC values. An algorithm called SG (Simulated annealing and Genetic) was developed to facilitate the optimal calculation of MIC, and the convergence of SG was proved based on Markov theory. When run on fruit fly data set including 1,000,000 pairs of gene expression profiles, the mean squared difference between SG and the exhaustive algorithm is 0.00075499, compared with 0.1834 in the case of ApproxMaxMI. The software SGMIC and its manual are freely available at http://lxy.depart.hebust.edu.cn/SGMIC/SGMIC.htm.

Ejemplares similares