I have reads with Barcodes in sequence and the sequencing facility told me, all the adapters are removed, but the fastqc report shows me, that they are not. Since Illumina Universal adapters and polG
are detected in the forward and the reverse reads
And then tried to remove the remaining adapters with cutadapt too, after some playing around I was able to remove the Illumina adapters and polyG with this command but only in the forward reads!
I tried with the original, the reverse, the reverse compliment and the compliment version of the Illumina universal adapter sequence for p-anywhere-r. But nothing worked to remove the adapters from the reverse reads.
These are the multiqc plots of the demultiplexed reads before and after trimming.
I have reads with Barcodes in sequence and the sequencing facility told me, all the adapters are removed, but the fastqc report shows me, that they are not.
They said that the barcodes were removed or that the adapters were removed? Or are you using those two terms to refer to the barcodes? They removed the adapters (not barcodes) but did not demultiplex?
I'm not familiar with the figures you shared. They show that an adapter sequence that you told the tool to look for is in your reads? Do the illumina adapter sequences that you shared come from the library prep protocol or somewhere else?
thank you for the quick repsonse and sorry for being not clear on that.
I do distinguish between adapters and barcodes. Adapters being the pieces required and attached by the sequencing facility and barcodes being part of my primer to distinguish my samples.
Yes exactely, they told me they removed the adapters and I only have to demultiplex them. But when I checked the demultiplexed reads with fastqc I got a red warning for all of the reads that they do have adapter content.
The figures were only supposed to make clear, that the adapters really only remain in the reverse reads after trimming.
Originally I looked at the fastqc report and these told me that the adapter sequences found are Illumina universal adapters. So I googled the sequences of these because I coulnd´t get ahold of the sequncing facility. And eventually I found the sequence which is within the code above (--p-anywhere-f AGATCGGAAGAGCACACGTCTGAACTCCAGTCA).
The only problem left is, that I don´t know what sequence, code parameter or else to use to get rid of these adapters in the reverse reads.
Here are the pictures of the fastqc output for only one sample (R1 and R2) before and after trimming with cutadadapt: R1 before timming:
Is the following correct? Fastqc (this is the software you're using right?) has a built in set of adapter sequences that it searches for by default, and it found the "Illumina Universal Adapter" and "PolyG" sequences in your forward and reverse reads. It doesn't tell you what the actual sequence is, so you searched online for them and what you found is what you provided to cutadapt. Cutadapt successfully removed the adapter sequences in the forward reads but failed to remove the adapter sequences in the reverse reads.
If this is the case I think it's safe to say that you're simply searching for the wrong adapter sequence in the reverse reads (whatever you found online was incorrect).
I know you said that you couldn't get ahold of the sequencing center, but I would try to again because it really is their responsibility to provide this information. Thanks to their negligence your analysis is stumped before you could even get started.
Otherwise, maybe trial and error some more adapter sequences. From a quick google search for "fastqc illumina universal adapter" there seem to be different answers as to which sequence(s) are actually used.
Hi @lea.k
Typically in our lab, we use tools such as BBDuk or Trimmomatic for adapter removal which can iterate through a list of adapter sequences (Not sure, if this can be done with cutadapt). Once this process is complete, we import the filtered reads into QIIME2.
I recommend giving BBDuk a try on the demultiplexed reads (export or unzip demultiplexed-seqs-V45-0.25.qza), as it includes a predefined list of adapter sequences.
In the predefined list, I was able to locate only one of the adapter sequences you mentioned: "AGATCGGAAGAGCACACGTCTGAACTCCAGTCA,". The other one was not present. As mentioned above, this could be the reason adapters were not removed from the reverse reads.
Thank you for the quick response and sorry for my late response, I was on holidays
That is correct. I assumed, since only in the reverse reads the removal didn´t work, I might just have to change some parameters in cutadatpt or change the sequence in reverse compliment or a combination of both and therefore thought maybe someone in the community knew wich combination could work.
But of course, maybe I have to try different adapter sequences. If so, do I have to alter the sequences or the parameters, since I am looking for these sequences in the reverse reads only?
I will try to get in contact with the sequencing facility too!
Thank you for the tip! I have tried trimgalore wich to my knowledge also goes trough a list of Nextera, Illumina, etc. adapters but was not successfull in removing them.
I will give BBDuk a shot though, thanks for all the additional info and the code!
