how to extract UMI barcode from .FASTQ file(GSM3139597)?

Hi, I need to remove degenerate sequence (N4) at the 5’ and 3’ of the reads. my target is GSM3139597. So for extracting UMI from .fastq, I install UMI tools. Now, I have to extract GSM3139597 from UMI barcode. based on instruction, for the single-end read, I have to use the below code:

umi_tools extract --extract-method=string --bc-pattern=[PATTERN] -L extract.log [OPTIONS].

I need a guide for handling GSM3139597 as input in that code. In other words, I don’t know how can I use that code for solving my problem. I don’t know how can I prepare arguments of that code relate to my GSM?

[IanSudbery]

I’m not sure I see any conflict. The post to the umi-tools documentation is instructions for how to craft a regex to deal with any library. The instructions for the NEXTFlex library are for that specific library and were written without any knowledge of UMI-tools.

You should process the reads to trim the 3’ adaptor sequence using your favourity adaptor trimmer. For example cutadapt is used by the the depositors of your GSE. (This is step one in the process outlined above).

The second step of the process in the NEXTflex data analysis section is to trim the first and last 4 bases of the adaptor clipped reads. In this protocol those bases are discarded. However, it looks like the depositors of the GSE you quote realised that as these bases were random bases, they could be used as a UMI, and used UMI-tools to remove them from the ends of the read and put them on the read name. To do this it is neccessary to tell UMI-tools that the “UMI” is the first and last 4 bases of the read.

In regex we specify any 4 bases as .{4}. In UMI-tools we specify bases to be captured by creating a named capture group, so a named capture group that captures 4 bases as the first part of a UMI would be (?P<umi_1>.{4}). We start our pattern with this to specify that we want to capture the first 4 bases of the read. If we wanted to make sure this was anchored at the start of the read, we can make doubly sure by adding a ^ to the start. ^(?P<umi_1>.{4}) We then need a similar capture group at the end of the read to capture the last 4 bases of the read as the second part of the UMI (?P<umi_2>.{4}). We can make doubly sure that is anchored at the end of the read by adding a $: (?P<umi_2>.{4})$. All we need now is to connect the the two together by any number of other bases to give:

^(?P<umi_1>.{4}).+(?P<umi_2>.{4})$

Interestingly while the depositors of the GSE use UMI-tools to grab the first and last 4 bases of the read as a “UMI”, they don’t appear to actually use that UMI for anything in their downstream pipeline. Unless they have an umi_tools dedup step that they have forgotten to include in their methods.

[TomSmithCGAT]

I’m closing this due to low activity. Please re-open if this issue still needs to be resolved