adapter trimming - help

[a_mt]

Hello all,

I am a newbie to NGS analysis. I recently got raw sequenced data (Illumina) for yeast.
Read length is 104 bps. I have average knowledge in R. I used ShortRead package for initial processing of reads. However I have some doubts/queries regarding adapter trimming. I searched many threads here and came to know about lot of trimming tools including these. According to many Trimmomatic found to be the best. Now I want to use this tool on my data. But this tool requires a fasta file (The Adapter Fasta) containing adapter sequence. Now,

1. How to know the adapter sequence??
2. My data is paired end reads, in this case how do I proceed??

How do I create this file?

I know my questions are too lame/simple for this forum and I am extremely sorry for such noob questions.

Help me.

Thank you.

[gaffa]

In the best case scenario, the people doing the sequencing experiment will inform you about the sequence of the adapters used during the experiment. However this is not always the case. What you can do if you do not have any information about the adapter sequence is to run a program like FastQC (http://www.bioinformatics.babraham.a...ojects/fastqc/) which will search your fastq files for a number of commonly used adapters. If your sequencing experiment was performed according to some standard protocol then the adapter sequence might very well be included in FastQC's list and if there is substantial adapter contamination in your data then this will be seen in the program report.

[a_mt]

Hello Sir,

Thank you for the reply. I randomly extracted 1000 reads (fastq) and ran Fastqc on it.
fastqc reported kmer contamination as fail and gave around 200> 5bp sequences.
What do I do now?? Do I have consider all of them as adapter sequnence.

And my data has come with this file barcodes.txt

Control_1_1 CACTGT
Control_1_2 ATTCCG
Control_1_3 GCTACC
Control_1_4 CGAAAC
Mutant_3_1 GATGCT
Mutant_3_2 AGCTAG
Mutant_3_3 GGCCAC
Mutant_3_4 ATTATA

are these adapters??

Again sorry for noob question!

Thank you.

[gaffa]

You should probably use much more than 1000 reads for FastQC - the program is pretty fast, I would use the whole fastq files if they are not enormous. What you can look at in the FastQC output is the "Overrepresented sequences" section, rather than the k-mer content (which is harder to interpret). If there is adapter contamination the adapter sequences should probably show up here, and if its a sequence which is present in FastQC's list of common adapters the identity will be listed in the "Possible Source" column.

And these barcode sequences are most likely the sample barcodes used for multiplexing (putting multiple samples into the same sequencing lane), and are not adapter sequences.

[a_mt]

I ran fastqc on larger fastq (>1M reads), and "there are no overrepresented sequences".
Now will it be alright if I continue with the alignment to a reference genome??

Thank you.

[TonyBrooks]

You should probably use much more than 1000 reads for FastQC - the program is pretty fast, I would use the whole fastq files if they are not enormous. What you can look at in the FastQC output is the "Overrepresented sequences" section, rather than the k-mer content (which is harder to interpret). If there is adapter contamination the adapter sequences should probably show up here, and if its a sequence which is present in FastQC's list of common adapters the identity will be listed in the "Possible Source" column.

And these barcode sequences are most likely the sample barcodes used for multiplexing (putting multiple samples into the same sequencing lane), and are not adapter sequences.

Won't that just look for adapter-dimer rather than adapter read-through?
You will need to ask who prepared the libraries for the adapter sequence.

[a_mt]

Hello sir,

apparently our samples were "outsourced" for sequencing and they have not given me the adapter sequences. I have mailed them regarding the same. Again thank you very much for the suggestions.

I must say Seqanswers - hats off to you!! I am learning so much from this forum.