What can we learn from the FastQC report?

[kipkurui]

View the HTML summary of the QC report and tell us what it means, what do we need to do based on the results?

[simeonhebrew]

One thing that stands out is a low quality portion of the reads in the first approximately 40 base pairs, the same region that has been reported to have a high percentage N-count.
We could consider trimming out this region.

[nanjalaruth]

From Ruth&Gatua

During trimming, the minimum length could be set to 50 and the phred quality score set to30.
Warning: Our data is characterized by low per base sequence content, abnormal GC content across the genome and a majority of the sequences are overrepresented and duplicated.
Mitigation (Things that could have been done):
Overrepresentation - normalization during library preparation
Study the genome to be able to know whether its characterized by repeats and whether it's AT rich or GC rich

[Kauthar-Omar]

From Kauthar and Rose

• Some sequences are contaminated with adapter sequences and removing the adapters will improve the quality score of the sequences.
• The thing to note is that our data has has abnormal GC content, low per base content, high sequence duplication and high levels of overrepresented regions.
  With the consideration that this is an RNA Seq library, the overrepresentation may be due to very abundant transcript as opossed to the normal conclusion of PCR enrichment bias.

[Asatsa]

From Asatsa and Wilson

We need to normalize the reads due to over-representation

There is high duplication, therefore dereplication needs to be done maybe using tools such as Mark Duplicates?..