Quantify gene expression using flanking sequences

Question

[Hana]

I am trying to quantify the levels of a gene (transduced) by conducting a grep search on both the 5' and 3' flanking regions on either side of the gene within my construct. However I see that there is quite a large discrepancy in read counts when comparing reads that contain the 5' and 3' sequences. Is it possible to discern which of the flanking sequences are individually present beside the gene (ex/ only has the 5' and no 3') or reads that have both flanking sequences beside the gene? Both sequences are around 40bp and each read in my fastq file is 75bp.

[michael.ante]

Hi Hana,

There are a couple of pipelines available for performing quantification in the way you described and as far as I can remember, Cufflinks was able to extend a gene's coordinates.

If you see a 5' or a 3' bias, it can be due to your sequence technology and retained primer sequences. In this case, you may need to give a bit more details about your experimental set up.