How to handle contamination in 16s amplicon data

Question

[1769mkc]

The sample of origin is csf fluid, which is supposed to be sequenced. Now running those sample in gels there are multiple bands for various samples, ideally it should be single band. Now if the libraries are sequenced and there would be contamination, what are the steps to clean the contamination before running the data analysis pipeline.

The pipeline I intend to use is DADA2 workflow.

Any suggestion or help would be really appreciated.

Answer 1

[GenoMax]

Now running those sample in gels there are multiple bands for various samples, ideally it should be single band.

If that is the case then you have no idea what is in there. If this is a non-replaceable sample (CSF fluid, so perhaps the case) then you may need to go on as is. This may be a "destined to" (or already) failed experiment.

You could filter against a 16S database after the sequencing (NCBI makes a representative 16S blast+ database available that you can extract the sequences from) and go from there.

[1769mkc]

You could filter against a 16S database after the sequencing (NCBI makes a representative 16S blast+ database available that you can extract the sequences from) and go from there. this will try

[GenoMax]

To add some clarity : filter as in use bbduk.sh from BBMap suite in filter mode to pull out reads that match 16S sequences (from NCBI 16S blast or other 16S databases).