1. Transcriptional Control and Microarrays


가정: 발현패턴이 비슷한 유전자는 발현 조절 기작이 같을 가능성이 높다.

 ※ Scheme (by George M. Church group at Harvard)




 Microarray Expression data

cluster #1

cluster #3

cluster #2

Motif Search (Gibbs Sampler Algorithm)








Upstream sequene retrieval

Motif 1

Motif 2

Motif 3











Clustering (k-means algorithm)

Construct transcriptional control network


The above scheme works quite well. (See Tavazoie et al., Nature Genetics 22, 1999)


We are trying to do these...

1) For this scheme, the microarray data used are mostly time series data. But k-means algorithm considers only the    similarity in expression patterns and does not have a sense of Time. We want to develop a clustering method that    better reflects the temporal patterns in the expression data. It will more reliably reconstruct the transcriptional    control network.

2) Extension of this scheme to the microarray data of species other than the yeast. This will include the development    of Gibbs Sampler alignment program optimized for the species.


Gibbs Sampler and multiple alignment ( a brief review)

Gibbs Sampler C code (can be downloaded at ftp://ftp.ncbi.nih.gov/pub/neuwald/ , Lawrence et al., science, 1993 )

George M. Church Lab Homepage(AlignACE, CompareACE can be downloaded)



Lawrence, C.E., Altschul, S.F., Bogouski, M.S., Liu, J.S., Neuwald, A.F., and Wooten, J.C. (1993), "Detecting Subtle Sequence Signals: A Gibbs Sampling Strategy for Multiple Alignment," Science, 262, 208-214.

Tavazoie, S., Hughes, J.D., Campbell, M.J., Cho, R.J., and Church, G.M. (1999) Systematic determination of genetic network architecture. Nature Genetics 22:281-5.

Hughes, J.D., Estep, P.W., Tavazoie, S. and Church, G.M. (2000) Computational identification of cis-regulatory elements associated with functionally coherent groups of genes in Saccharomyces cerevisiae. J. Molec. Biol. 296: 1205-1214.