Hello everyone. I am using limma for microarray analysis for some time. Now I am trying to use mergeScans function to merge scans made with different laser intensities. Each time I get the same error. I have tried to google- out the solution but I do not much understand the discussions I've found. As I am just a biologist and my R activity is mostly modifying commands that I have found in the manuals or web., my question is - is there any simple way to deal with it, without large programming skills and knowledge (plese explain what kind of error is that, is it something wrong with my data, what can I do) or do I need to find someone who knows R better than me to help me overcome it? Thank you Agnieszka This is one of my trials - Agilent two-color arrays, scanned and analyzed in GenePix: > RG1=read.maimages(targets1, source="genepix.custom", other.columns=c("ControlType"), wt.fun = wtflags(weight=0, cutoff=-50)) Custom background: MorphologicalOpening Read 252162310143_1_of_0004_sat1_0635.gpr > RG2=read.maimages(targets2, source="genepix.custom", other.columns=c("ControlType"), wt.fun = wtflags(weight=0, cutoff=-50)) Custom background: MorphologicalOpening Read 252162310143_1_of_0004_0635.gpr > > RG1 <- mergeScansRG(RG2,RG1) Error in nls(y ~ .hockey(x = x, alpha1, beta1, beta2, brk), start = list(alpha1 = alpha1, : step factor 0.000488281 reduced below 'minFactor' of 0.000976563 .pl

Not answering your specific questions, but an alternative is to use: library("aroma.light"); r <- calibrateMultiscan(R); g <- calibrateMultiscan(G); where R is an JxK matrix holding the K scans of the J (foreground) signals in the red channel. The calibrated/merged output is a Jx1 matrix. Likewise, G is a JxK matrix for the green channel with the merged signals in 'g' (a Jx1 matrix). Use foreground signals only - do not use background corrected signals. There is no need for tuning parameters. It works out of the box (unless "funny" things were done during scanning). See help("calibrateMultiscan.matrix", package="aroma.light") for further help and explanation. Further details, figures and results can be found in: H. Bengtsson, J. Vallon-Christersson and G. J?nsson, Calibration and assessment of channel-specific biases in microarray data with extended dynamical range, BMC Bioinformatics, 5:177, 2004. http://www.biomedcentral.com/1471-2105/5/177/ /Henrik 2012/1/26 Agnieszka ?mie?ko <akisiel at="" ibch.poznan.pl="">: > Hello everyone. > > I am using limma for microarray analysis for some time. Now I am trying to > use mergeScans function to merge scans made with different laser > intensities. Each time I get the same error. I have tried to google- out the > solution but I do not much understand the discussions I've found. As I am > just a biologist and my R activity is mostly modifying commands that I have > found in the manuals or web., my question is ?- is there any simple way to > deal with it, without large programming skills and knowledge (plese explain > what kind of error is that, is it something wrong with my data, what can I > do) or do I need to find someone who knows R better than me to help me > overcome it? > > Thank you > > Agnieszka > > This is one of my trials - Agilent two-color arrays, scanned and analyzed in > GenePix: > >> RG1=read.maimages(targets1, source="genepix.custom", >> other.columns=c("ControlType"), wt.fun = wtflags(weight=0, cutoff=-50)) > Custom background: MorphologicalOpening > Read 252162310143_1_of_0004_sat1_0635.gpr >> RG2=read.maimages(targets2, source="genepix.custom", >> other.columns=c("ControlType"), wt.fun = wtflags(weight=0, cutoff=-50)) > Custom background: MorphologicalOpening > Read 252162310143_1_of_0004_0635.gpr >> >> RG1 <- mergeScansRG(RG2,RG1) > Error in nls(y ~ .hockey(x = x, alpha1, beta1, beta2, brk), start = > list(alpha1 = alpha1, ?: > ?step factor 0.000488281 reduced below 'minFactor' of 0.000976563 > > .pl > > _______________________________________________ > Bioconductor mailing list > Bioconductor at r-project.org > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: > http://news.gmane.org/gmane.science.biology.informatics.conductor