Difference between revisions of "HughesLab:JTK Cycle"

From OpenWetWare
Jump to: navigation, search
Line 26: Line 26:
[http://www.umsl.edu/~biology/ Department of Biology]<br>
[http://www.umsl.edu/~biology/ Department of Biology]<br>
University of Missouri, St. Louis<br>
University of Missouri, St. Louis<br>
(Starting in August, 2013)

Revision as of 06:02, 30 May 2013


Home        Lab Members        Research        JTK_Cycle        CircaDB        MuscleDB        Publications        Protocols        Courses        Academic Analytics        Open Positions        Contact       


Figure 2, Hughes et al. (2010)
ROC curves showing that JTK_Cycle out-performs COSOPT and Fisher's G-test.

JTK_Cycle is a non-parametric algorithm developed in collaboration with Professor Karl Kornacker (Ohio State University) and Professor John Hogenesch (University of Pennsylvania School of Medicine). Its purpose is to identify rhythmic components in large, genome-scale data sets and estimate their period length, phase, and amplitude.

JTK_Cycle runs in R, a statistical language that can be downloaded here. The scripts necessary to run JTK_Cycle are available as a .zip: File:JTK Cycle (2-19-13).zip. This compressed file also contains a User's Guide and several example data sets. One important note: this is version 2.1 which includes improvements in the accuracy of amplitude calculations as well as confidence intervals for amplitude.

When publishing data analyzed by JTK_Cycle, please cite Hughes et al. (2010) Journal of Biological Rhythms. If you use the recently modified JTK_Cycle script to compare the confidence intervals of amplitude measurements, please also cite Miyazaki et al. (2011) Plos ONE.

If you have any difficulties getting JTK_Cycle to work on your data or need help with interpreting the results, please don't hesitate to ask: michael.evan.hughes <at> gmail.com.

UMSL logo.jpg

Hughes Lab
Department of Biology
University of Missouri, St. Louis