I'm a user of the no-longer existing Stanford Microarray Database (SMD), and very new to MeV. We scan Agilent 4x44k gene-expression 2-color arrays using a Genepix 4000b scanner. After gridding the array in the Genepix software (Axon), we get a ".gpr" result file. What we used to do was load this (in combination with the image files in .tiff) in SMD and proceed to data filtering: select features for which intensities are 1.5x over background in both colors, features that have a 3-fold change over the mean, and for which the values are found in 75% of the samples. This would give us a preculstering file, a ".pcl" file. We could then take this .pcl file and load it into Cluster/TreeView.
When I load multiple .gpr files in MeV using the file loader specific for GenePix files, I have no idea how the Cy5/Cy3 ratios displayed are calculated (background values removed from the feature value? A 1.5x factor over background for each color?). What columns of the .gpr files are used?
I saw I can save the matrix file, it looks like a preclustering file. Is there a way in MeV to proceed to filtering with .gpr files as I mentioned above? I saw how to do for 75% of present data over the samples (percentage cutoff filter), but the 1.5x over background and 3-fold change, I can't find where…
Thank you for your help. Sincerely,
Log in to post a comment.