Biocomputing and Media Research Lab

Image: Figures pertaining to research

Phenotype Mining and Analysis


Studying the genetic control of molecular, anatomical and/or morphological phenotypes in model organisms is a powerful tool in the functional analysis of a gene. The goal of our research is to develop algorithms that discover phenotypes of behavior in model organisms, which may identify, categorize, and quantify these phenotypes under conditions of minimal a priori information. Towards this, we are specifically exploring algorithms for motion-pattern discovery, recognition, and quantification in digital video. This can, for instance, allow discovery of phenotypic manifestations that tend to occur over longer periods of time than even a trained human might have the ability to detect. Furthermore, the algorithmically identified phenotypes are non-subjective and rigorously quantifiable.


  1. A. Shimode, I. Yoon, M. Fuse, H. C. Beale, and R. Singh, "Automated Behavioral Phenotype Detection and Analysis Using Color-Based Motion Tracking", Canadian Conference on Computer and Robot Vision, pp. 370 – 377, 2005. [PDF]
  2. A. Vaughan, R. Singh, A. Shimoide, M. Fuse, I. Yoon, "EigenPhenotypes: Towards an Algorithmic Framework for Phenotype Discovery", Proc. IEEE Computational Systems Bioinformatics Conference (CSB), 2005 (Poster Paper). [PDF]
SF State Home