Systemix Institute

Chlamydomonal reinhardtii is a green alga with genetics similar to yeast, but it also contains flagella that are virtualy identical to human cilia. In this project, we have conducted a genome-wide analysis of mRNA transcript levels during flagellar regeneration in Chlamydomonas reinhardtii using oligonucleotide microarrays. Homologs of several important human ciliary disease-related genes are identified by this study.

Reference: Stolc V, Samanta MP, Tongprasit W, Marshall WF. "Genome-wide transcriptional analysis of flagellar regeneration in Chlamydomonas reinhardtii identifies orthologs of ciliary disease genes." Proc Natl Acad Sci USA (2005) 102(10):3703-7. Download

1. Supplementary materials
2. Locations of genes and signals
3. Plot signal anywhere of the genome
4. Interesting genes
5. Filtering of expressed genes
6. Motifs

Download supplementary materials
Table S1 Genes used for validation of microarray data
Table S2 Genes that are upregulated 100% or more
Table S3 Average expression level for all predicted genes

Genes and signals

For probe selection, first the locations of all predicted genes were determined on the genome sequence available at the JGI website. Three different methods were used to complete this task. First, custom softwares were used to match the genome sequence and the protein sequence. The results of this search are in the 'highest confidence' file. Next, BLAT programs developed by Jim Kent of UCSC were used. The results are in the 'medium confidence' file. Finally, attempts were made to locate the remaining few proteins using BLAST and other methods. They are in the 'low confidence' category. In total, genetic locations of 19803/19832 predicted genes were determined.

Probes were selected for each exon of each mapped gene and then few additional probes were selected for the large intergenic region. Hybridization intensities for these probes were measured at 0-, 30, 45 and 120 minutes after removal of the flagella. Average signal for each gene is the median signal of all the constituent probes.

Highest confidence	file-1	15400 proteins	signal, percentage format
Medium confidence	file-2	15000 proteins	signal, percentage format
Low confidence	file-3	about 2560 proteins	signal, percentage format

Data format: column 1: protein name, column 2: scaffold name, column 3: map of protein on
scaffold, column 4: other information

In case of overlap between file-1 and file-2, use information in file 1.

Plot hybridization signals

View by gene ID

JGI ID

log scale

or by location

Scaffold		Strand
		CW
Begin		End
t=0	t=30	t=45	t=120
log scale

Interesting genes

These genes are linked with flagellar regenaration or human ciliary diseases byprevious studies.

Interesting genes:	list, regulation, html
Important genes:	list, regulation, html
688 genes from the Cell paper of Li et. al.	here

Filtering of strongly induced genes
i) Filter out all genes that do not have maximum deviation from t=0 atleast 20% at either t=30 or t=45.

ii) Filter out all genes that do not have minimum deviation from t=0 at least 5% at either t=30 or t=45.

iii) Filter out all genes that whose deviation at t=120 do not fall to less than .6 times maximum deviation at t=30 or t=45. Genes left after applying this filtering criteria are here.

More generalized:

MAX(%): MIN(%): FINAL/MAX:

Motifs
MDSCAN and MEME were run on 500-base upstream regions of 122 mostupregulated genes. Our method searched all 8-13 base words against 500-base upstream regions of all genes to find whether any of thosewords were over-represented in the highly upregulated genes.

MDSCAN	8 base motifs, 9 base motifs, 10 base motifs
MEME	MEME results, MAST results
Our method	results