Hi Jari, On Jan 2, 2008, at 3:11 PM, Jari Häkkinen wrote:
> Collett, James R skrev: >> Happy New Year, BASErs! >> >> We are developing metagenomic assays using Affy chips with >> custom-designed probe sets. The tools available for analyzing >> standard >> Affy gene expression data using pm-mm comparisons, however, are not >> relevant to our chip design and analytical approach (we're only >> using pm >> probes). >> > > We are aware of the shortcomings of the current plug-in. > Unfortunately, > we currently have no time to extend and improve it. Another reason > that > further slows down the progress is that we do not use the Affymetrix > platform ourselves. There are many improvements that could be done as > you point out. There are users who extend the affymetrix plug-in but > these additions have not become public yet and I am not sure if they > will be published. > One of these users would be me, I guess. I indicated that an extension was my plan some time ago, but unfortunately haven't had the time to complete it during the fall. I am now in the progress of completing the plugins and would really appreciate some help regarding how to compile the package. I have created two plugins: APTGeneotypePlugin and APTSummarizePlugin that implements an abstract plugin called APTPlugin. This abstract plugin extends the abstract plugin called AbstractExternalBinaryPlugin in the se.lu.thep.affymetrix package. For now, I have created a new package called APT. This package includes a copy of AbstractExternalBinaryPlugin.java. I have tried to include the affymetrix.jar file in the Makefile to keep it separated from my source code. This, however, fails. I don't really know how to do it nicely. What do you think would be the best way to proceed? I could either extend the existing affymetrix package with my two new plugins or create a separate package as described above. The first option is perhaps best if James Collett is to add new plugins based on e.g. apt- cel-extract. On the other hand, the very same functionalities in the Plier and RMAExpress plugins are given by APTSummarizePlugin. Another question is what to call the package. Now it is called se.uu.lcb.APT, but that does not seem fair as I have reused some of the code in Plier.java. What are your suggestions? Thanks, Robin > >> Accordingly, we would like to adapt the BASE affymetrix.jar plugin to >> use other executables in the Affymetrix Power Tools suite besides >> just >> the apt-probeset-summarize. Specifically, we would like to use >> apt-cel-extract within affymetrix.jar to simply extract feature >> intensities, probe IDs, probe set IDs, and XY coordinates from CEL >> files >> into new root bioassay sets. I have already used apt-cel-extract >> on the >> command line to successfully extract these fields from CEL files into >> text files. > > It should probably be straightforward to add more functionality to the > plug-in using the different Affymetrix tools. In the simple case one > would just add an option for the user to select which tool to use, > and a > more fancy way would be that the plug-in checks which tools are > appropriate for the data at hand. > > >> For now, simply trying to feed our CEL files into the current Affy >> PLIER >> plugin within BASE 2.5.0 gives (not surprisingly) the following >> error: >> >> java.io.IOException: Unexpected exit of PLIER sub-process. Return >> value: >> 1 >> at se.lu.thep.affymetrix.Plier.runBinary(Plier.java:157) >> at >> se.lu.thep.affymetrix.AbstractExternalBinaryPlugin.run >> (AbstractExternalB >> inaryPlugin.java:262) >> at >> net.sf.basedb.core.PluginExecutionRequest.invoke >> (PluginExecutionRequest. >> java:89) >> at >> net.sf.basedb.core.InternalJobQueue$JobRunner.run >> (InternalJobQueue.java: >> 435) >> at java.lang.Thread.run(Thread.java:619) >> >> Are any of you interested in working with us to expand >> affymetrix.jar to >> include functionalities from apt-cel-extract and perhaps others in >> the >> the Affy Power Tools suite? If not, can you give us some >> suggestions on >> how we might best get started on this, and what kinds of >> challenges or >> incompatibilities that we might face? > > This is a great idea! The different apt tools should be in one single > package. This will make Affymetrix usage simpler for BASE users > (and the > administrators). So, we are happy to work with you on improving the > plugin. > > The first steps forward. You already have read access to the source > code > for the plug-in through > http://baseplugins.thep.lu.se/wiki/se.lu.thep.affymetrix Read the > instructions on this page on how to check-out the latest > development code. > > >> Thanks, and Congratulations on BASE 2.5.0 - it's been working >> exceedingly well for us! > > Thank you, > > Jari > > >> Jim Collett >> >> James R. Collett, Ph.D. >> Systems Biology Fellow >> Chemical and Biological Process Development Group >> Pacific Northwest National Laboratory > > ---------------------------------------------------------------------- > --- > This SF.net email is sponsored by: Microsoft > Defy all challenges. Microsoft(R) Visual Studio 2005. > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > _______________________________________________ > basedb-devel mailing list > basedb-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/basedb-devel --- Robin Andersson, PhD student Linnaeus Centre for Bioinformatics Uppsala University BMC, Box 598, SE-751 24 Uppsala, Sweden phone: +46 18 471 66 86 (admin: 66 90) fax: +46 18 471 66 98 office: BMC, C6:315a email: [EMAIL PROTECTED] http://www.lcb.uu.se/~robina/ ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ basedb-devel mailing list basedb-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/basedb-devel
