Fastq with very poor per tile quality but excellent perbase quality - is this a tampered fastQ?

Question

[Arun]

We recently got a set of targeted genes sequenced(probably they used novaseq 6000). The panel is an extended brca panel for hereditary breast cancer screening. However after initial checks in FastQC, we observed that the per tile quality score for one read pair is quite bad but the per base quality parameter is extremely good. Does this signify tampering of quality values ?

The issue is with one pair of paired end reads. check the image below

find the FastQ here https://easyupload.io/b9yvwn https://easyupload.io/jsch1c

[GenoMax]

Perhaps this has something to do with failure on Adapter content. What does that plot look like.

In general, you should analyze the data and if there are issues with the results then backtrack to see if you can identify the issue. This tile plot probably should not be a deterrent to do that.

[Arun]

Thanks GenoMax for the comment here is the adapter content part

another thing I noted was that the quality values in fastQ is made up of F, :, and ,. This is true for the entire length of FastQ. here is the screen shot

[Carlo Yague]

Reasons for seeing warnings or errors on this plot could be transient problems such as bubbles going through the flowcell, or they could be more permanent problems such as smudges on the flowcell or debris inside the flowcell lane. [SOURCE]

Here we can see both transient and more permanent drop in quality (that persists over several cycle). Something probably didn't work perfectly well during the sequencing run, but nothing to worry too much about IMHO. I don't see how this would be a symptom of data tampering.

[Arun]

Thanks @CarloYague! Another thing I noted was that the quality values in fastQ is made up of F, :, and ,. This is true for the entire length of FastQ. here is the screen shot