binning Illumina reads by clusters on flowcell

[apratap]

Just wondering is it possible to group the reads by the cluster they
came from on the flowcell. We want to do so to get an idea of regions(clusters)
contributing to low/high quality reads and compare GC.

I have noticed the fastq header has the X|Y coordinates but not sure how to cluster them together to determine the boundaries of each cluster.

Thanks!
-Abhi

[HESmith]

Hi Abhi,

A note on Illumina terminology: each cluster is a single read, so you must be referring to a group of clusters within a defined space.

If your data is from GA, then it was divided initially into 100 (or 120) tiles. IVC and error plots are generated for each tile. I suspect that's not sufficient resolution for your application. As you noted, you could bin the data by X-Y coordinates and plot average (or mean/min/max) scores to find regions of low or high quality. I don't know of any off-the-shelf tool for this purpose.

[apratap]

Hi Smith

Thanks for correcting me. I indeed meant cluster of clusters i.e a region on the flowcell. What I am not sure what size regions should I limit myself to.

I think what you suggest makes sense for starting and see how clustered my good and bad regions are in terms of quality and may be GC

-Abhi