I have to pre-process around 10,000 fastqQ files containing 16S rRNA sequences. According to the multiQC reports generated before pre-processing the sequences have good quality reads with Q score above 30 for 9000+ samples. However, in the case of sequence representations and sequence duplication levels most of the samples fail. But for adapter content more than 9000 pass.
Hello! You can use FastQC to find the adapters or overrepresented sequences (up to 50bp) and you can remove them with cutadapt (-a for single-end, -A for paired). I also like to do quality trimming with Trimmomatic (sliding window, or maxinfo). You can re-run FastQC in-between to see how in is improving. After trimmomatic I use STAR with an Ensembl genome (gencode if mouse or human) I conform, if possible, strictly to the ENCODE parameters listed in the STAR manual. Good luck!!
The log files of FastQC (maybe multiQC too?) Will have the sequences. Just copy them into excel and then the column for sequences into a text editor and supply them (or just determined adapter sequence) to cutadapt which you can install with "pip3". Hope it works out!
- Badges
- Science topic
- Similar topics
- Analytical Chemistry
- Extraction