|
Status |
Public on Sep 25, 2010 |
Title |
Arb1_wt_1 |
Sample type |
genomic |
|
|
Channel 1 |
Source name |
DamArb1_SPB382_1
|
Organism |
Schizosaccharomyces pombe |
Characteristics |
strain: SPB382 expression: Dam-Arb1
|
Treatment protocol |
Harvested cells were spheroplasted to remove cell wall
|
Growth protocol |
Strains grown at 30 degress in YES medium to OD600=0.4
|
Extracted molecule |
genomic DNA |
Extraction protocol |
Genomic DNA isolated using Dneasy Blood and Tissue Kit (Qiagen)
|
Label |
biotin
|
Label protocol |
DNA was fragmented and labeled with the GeneChip Whole Transcript Double-Stranded DNA Terminal Labelling Kit (Affymetrix)
|
|
|
Channel 2 |
Source name |
Damonly_SPB492_1
|
Organism |
Schizosaccharomyces pombe |
Characteristics |
strain: SPB492 expression: Dam only
|
Treatment protocol |
Harvested cells were spheroplasted to remove cell wall
|
Growth protocol |
Strains grown at 30 degress in YES medium to OD600=0.4
|
Extracted molecule |
genomic DNA |
Extraction protocol |
Genomic DNA isolated using Dneasy Blood and Tissue Kit (Qiagen)
|
Label |
biotin
|
Label protocol |
DNA was fragmented and labeled with the GeneChip Whole Transcript Double-Stranded DNA Terminal Labelling Kit (Affymetrix)
|
|
|
|
Hybridization protocol |
hybridization was performed for 16h at 45°C and 60rpm in an Affymetrix hybridization oven; washing steps were done in an Affymetrix Fluidics station with the Affymetrix protocol FS450_0002
|
Scan protocol |
GeneChips were scanned with Affymetrix GCC Scan Control v. 3.0.0.1214 on a GeneChip® Scanner 3000 with autoloader.
|
Description |
Not used in publication due to poor reproducibility, likely resulting from non-functionality of Dam-Arb1 fusion protein
|
Data processing |
oligo-level data was rma background corrected, quantile normalized and log2 transformed using Bioconductor packages tilingArray and preprocessCore on R version 2.9 expr_M_CorV2.wig wiggle file showing the results of all the above contrasts. Normalization of the tiling array data: All tiling arrays were processed in R using bioconductor and the packages tilingArray and preprocessCore. The arrays were RMA background corrected, quantile normalised and log2 transformed on the oligo level using the following command: expr <- log2(normalize.quantiles(rma.background.correct(exprs(readCel2eSet(filenames,rotated=TRUE))))). Contrasts were computed on the oligo level by subtracting respective columns of expression. Reannotation of the Affymetrix S. pombe Tiling 1.0FR Array: The sequences of the 1174792 perfect match oligos were extracted from the BPMAP file Sp20b_M_v04.bpmap (www.affymetrix.com) using the readBpmap function from the affxparser package. Alignment to the S. pombe genome (May 8, 2009, http://www.sanger.ac.uk/Projects/S_pombe/) was performed by the software bowtie (version 0.9.9.1) allowing for up to 100 matches per oligo. Correcting a bias caused by variable fragment size between GATC restriction sites: While inspecting oligo level contrasts in a genome browser we noticed enrichment breakpoints coinciding with GATC restriction sites. Therefore we speculated that during sample preparation, certain fragment sizes might be depleted or enriched. Plotting the fragment sizes between two GATC restriction sites against the average enrichment in the corresponding fragment confirmed this observation on a genome-wide scale. We speculated that this resulted from small differences in the sample preparation process, probably during the step of size selection. In addition to fragment size-dependent mean enrichment/depletion, we noticed discontinuity of the contrast variability coinciding with GATC restriction sites. We plotted the fragment sizes between two GATC restriction sites against the standard deviation of the enrichment in the corresponding fragment and observed discontinuities on a genome-wide level. Both observed effects are highly unlikely to be of biological origin because they are tightly correlated to the fragment size between GATC restriction sites. Therefore, we designed a software that would at least partially reverse the bias and correct the initial data. For every contrast independently, the corrector walks through all the fragments (between GATCs) and adjusts the mean and the standard deviation of all the oligos that are located within the fragment. The adjustment is fragment size-dependent and the extent of correction is determined from the lowess fit of the corresponding contrast. In more detail, for a given fragment size, the value of the lowess smoother is evaluated and compared to the average level of the given contrast. Therefore the corrector does not modulate the overall mean or the variability of the contrasts.
|
|
|
Submission date |
Sep 24, 2010 |
Last update date |
Sep 24, 2010 |
Contact name |
Katrina Woolcock |
Organization name |
Friedrich Miescher Institute for Biomedical Research
|
Lab |
Bühler
|
Street address |
Maulbeerstrasse 66
|
City |
Basel |
ZIP/Postal code |
4058 |
Country |
Switzerland |
|
|
Platform ID |
GPL7715 |
Series (1) |
GSE24360 |
DamID for Swi6, Rdp1 and Dcr1 in Schizosaccharomyces pombe |
|