Strand-specific Library Types

[Phelelani Mpangase]

Dear all,

Im a trying to do a transcriptome assembly using Trinity. The paired-end reads I am working with were generated using the Illumina TrueSeq mRNA stranded kit.

When specifying --SS_lib_type, would I use FR or RF for the orientation if my reads? Is there a way of knowing what orientation my reads are in?

On the Trinity paper (Haas et al, 2013) as well as Trinity documentation, it says that the dUTP-based strand-specific sequencing generates ‘RF’ paired-end sequences.

Could someone shed some light on this for me.

Regards,

Phelelani Mpangase

[Ken Field]

Dear Phelelani-

The TruSeq kit is dUTP-based and generates RF orientation reads.

Ken

--
You received this message because you are subscribed to the Google Groups "trinityrnaseq-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to trinityrnaseq-u...@googlegroups.com.
To post to this group, send email to trinityrn...@googlegroups.com.
Visit this group at http://groups.google.com/group/trinityrnaseq-users.
For more options, visit https://groups.google.com/d/optout.

--

Ken Field, Ph.D.

Associate Professor of Biology

Program in Cell Biology/Biochemistry

Bucknell University

Room 203A Biology Building

[Phelelani Mpangase]

Dear Ken,

Thank you for your response.

Another question... Would I be right if I combine all the left reads (reads with a /1 suffix) into a single file (all_left_reads.fastq.gz) and all right reads (reads with a /2 suffix) into a single file (all_right_reads.fastq.gz), then running the fillowing Trinity command for the assembly?

Trinity --seqType fq --max_memory 24G --left all_left_reads.fastq.gz --right all_right_reads.fastq.gz --SS_lib_type RF --CPU 6

Phelelani

[Ken Field]

Yes, that is correct. Just be sure that there are no spaces in the header lines and that the only difference between the left and right pairs are the /1 and /2.

As a side note, 24G might not be enough unless you have under 100 million reads. And if you have over 100 million, normalization is highly recommended.

Good luck!

Ken

[Phelelani Mpangase]

Dear Ken,

I am working with about 50 million reads. It took about 8hrs for the assembly to complete using 24Gigs of memory and no normalization.

Thanks again for your help.

Phelelani

[T. T.]

May I ask a related question?

I am reassembling raw data files from NCBI (Illumina dUTP-based strand-specific sequencing). When I split the sra-file into two fastq files I get files which do already have the suffix 1 or 2. Does 1 now correspond to reverse and 2 correspond to forward? (And furthermore, I will assemble my own transcriptome soon, when I get back my data from the sequencing center (same method), I assume they will also be with suffix 1 and 2, right?)
When I am using Trimmomatic on them, the first input file results in 2 forward output files and the second input file results in 2 reverse output files, right? Thus, I need to write the inputfile with suffix 2 first right? Then I would concatenate the two forward output files (paired and unpaired) of all samples into my all-left file.

Do I still need to set SS_lib_type RF in Trinity?

Excuse me for this messy question but I get kind of confused about all the different names. Forward=left=sense ? Reverse=right=antisense ?

The code I would use for Trimmomatic is the following:

java -jar /path_to_trimmomatic/Trimmomatic-0.32/trimmomatic-0.32.jar PE -threads 2 -phred33 -trimlog trim52.log SRRXXXX52_2.fastq.gz SRRXXXX52_1.fastq.gz set52_forward_paired.fq.gz set52_forward_unpaired.fq.gz set52_reverse_paired.fq.gz set52_reverse_unpaired.fq.gz ILLUMINACLIP:/path_to_trimmomatic/Trimmomatic-0.32/adapters/TruSeq3-PE-2.fa:2:40:20 LEADING:5 TRAILING:5 SLIDINGWINDOW:4:15 MINLEN:36 2>set52_trimmomat.out

Thank you very much in advance.