|Data analysis||Translator Libraries|
|L2L Microarray Database||Code for Tools|
Interpreting the data you generate with L2L can be challenging. Our own experience is that the first step is determining whether you can trust the L2L results - statistical validation. Next is the biologically critical step of identifying interesting patterns in the data. Finally, we return to statistics to decide whether we can trust those patterns. Our own analyses of diabetic nephropathy and the progeroid Cockayne syndrome with L2L offer examples that you may find helpful to cite or follow. We are happy to offer collaborative help in replicating the steps in our own analysis with your data. This can be as straightforward as running a random-data simulation to validate the P values; or as complex as rigorously testing whole categories of gene lists against your data, as we did with Cockayne syndrome. We can also be helpful if you have a very large number of analyses to run and do not have the expertise to take advantage of the command-line L2L application. Feel free to contact us at any point in your data analysis process.back to top
1. Collaborations with other analysis tools and databases. We welcome collaborations and data-sharing with the developers of other bioinformatics tools. Please feel free to contact us if you believe your project and ours would benefit from sharing. Note that the L2L databases can be freely downloaded, are to free to use privately without restriction, and can be freely distributed with any GPL-compatible software tool. We do ask, if you use these data in your tool, that you acknowledge L2L in a place visible to users and cite our publication in any manuscript describing such a tool.
2. User-submitted data. If you found L2L useful, we ask that you consider submitting your own experimental data as lists for the database; this should take only minutes of your time, and will help make L2L more useful for the entire community. Use the form below to submit your data as a list of probes with a few brief annotations. We appreciate your help in continuing to grow the L2L database, making it more useful for all users.back to top
The lists that make up the L2L MDB are all in the form of gene symbols, to allow comparisons across different microarray platforms. The data you analyze, however, is in the form of platform-specific probe IDs, to preserve the greatest possible detail of your data for analysis. In order to translate between gene symbols and probe IDs, L2L needs a translator library for each microarray platform. This is a plain-text file that contains the probe ID and corresponding HUGO symbol for every probe on the microarray that is annotated with a gene association.
L2L includes translator libraries for a variety of Affymetrix, Agilent, Illumina, and NIH/NIA/JHU cDNA microarrays. Data from any of these patforms can be analyzed easily by L2L. For data from other platforms, the user must create their own translator library, and use the More Choices form to submit it with their data. If you need to make your own library, please consider contacting us about adding it to the L2L repertoire, so that other researchers using that platform do not need to duplicate your effort.back to top
All of the software associated with L2L, including the L2L application, the MammalHom tools, and the L2L-GO tools, are released under the GNU General Public License. The source code is therefore freely available (in fact, since they are written in Perl, the source code is what is we are distributing). If you have any suggestions for changes or patches to any of the code for any of the applications, please contact John Newman. We welcome all suggestions, and if your code is incorporated into a new release of L2L, we will offer proper acknowledgement, as well as our gratitude.
On a grander scale, we also welcome volunteers to create GUI-friendly verisons of the L2L application. We plan to create a MacOSX-native version of L2L, consisting primarily of a Cocoa GUI wrapper around the command-line L2L application. For other platforms (Windows and Linux, in particular), creating such a wrapper may be a very undertakable undertaking for anyone with some experience developing for that platform. If you might be interested in creating a GUI port, again, please contact John Newman.back to top