I took a picture of an Illumina plate (using QSEQ files)

Illumina sequencer output (QSEQ format) contains reads, quality scores and the X/Y coordinates from the cluster they came from.

I was curious if certain parts of an Illumina plate have lower quality reads than others. So, I decided to draw a map of an Illumina plate using these coordinates and the average quality score for that read.

In the first image, a white pixel corresponds to the presence of a sequence at a coordinate:

Black and White Illumina Photograph

You can see a subtle border between the eight lanes but not much else.

In the second image, I used two colors: Green for average quality scores above 99% (good) and Red for below (bad). This way, approximately half of the clusters are red and half are green.

Second image, in color:

Color Illumina Photograph

I did this using the Python Image Library and Python. You can see the code on GitHub.

I’m sure the Illumina software already does something like this but you have to buy an Illumina machine to get it.