eQTLViewer's role in eQTL mapping

eQTLViewer is a tool to plot (thousands of) expression QTLs after they are claimed through various statistical tools. Before making the 2D plot, users need make sure a few things:

  • A QTL is a genomic region, containing genetic markers or genes. Majority QTL mapping softwares (in the orange box above) will declare QTLs with respect to a genetic map. If there is a physical map available for the mapping population, do you want to arrange eQTLs according to the physical positions of markers/genes contained in the eQTLs?

  • How do you want to arrange your expression traits? In eQTL mapping, each trait is a gene (each eQTL is a consecutive group of genes on a chromosome). It is natural to arrange traits in the same way to arrange eQTLs. However, there might be biological reasons that you want to cluster certain traits together in all sorts of ways.

  • Given the popularity of biological databases available for (model) organisms and fantastic database integration/inter-connection works of these sites , the 2D plot generated by our eQTLViewer allows users to connect each genetic feature displayed on the picture to a database for the organism.

    • Due to uncertainties in methods/softwares used to claim QTL regions and various requirements of visualization, we ask user to provide all their answers to the above questions, as well as eQTL information, in an xml formatted input file . Following this link, you will see an XML file displayed using SVG plot. It happens to act correctly in FireFox 1.5. To view it in IE, remember to press 'Alt' key to scroll. The actually XML file can be found here.

    You will need knowledge of text processing languages (eg PERL) and rudimental understanding of the way to query database. Here is my PERL code as an example to prepare the xml input file. It reads .GFF file to get genomic position for each genetic feature. I will be very glad to help you in case you need.

    No biological annotation information stored in SVG

    The 2D plot generated by eQTLViewer can retrieve properly formatted biological annotation from external files (again, PERL programming is important here) and compare them with eQTL mapping results. For security reasons, SVG can only access files in the same server, or in your own machine after being downloaded (To download a svg file, right-click on the graph and choose 'Save SVG As...'). If you do not have such external files that are accessible to the downloaded SVG, the left pattern that appears in our complex examples will disappear.

    Technical details in providing annotation to SVG graph

    Upon loaded into browsers, the 2D plot will try to read a file called 'file.list' in the same directory. Based upon the information in this file, a left panel of annotation groups is constructed as in our complex examples

    file.list contains a list of files in the same folder, each of which corresponds to a type of biological annotation. For example, in our complex examples , one line in file.list is 'transcriptional_Factor chartreuse', while all known transcriptional factors-target pairs in yeast are stored in a file called transcriptional_Factor. Such pairwise relationship is directional in the sense that transcriptional factors are regulators of their targets; if we found for a transcriptional factor, its expression trait was mapped to its target, we do not think such a QTL is compatible with a regulation pathway we know. However, if the roles of transcriptional factor/target are reversed, we think there is a match between the QTL and previous biological knowledge. Such an eQTL will be marked out with chartreuse dot. This color is specified in file.list and is used to draw the transcriptional_Factor entry in our 2D plot. On the other hand, MIPS : pairwise relationship between genes in a same protein complex is not directional, which is also the case for KEGG : genes in a same pathway . Check file.list for more format information