Merge Paired-End Reads

Question

[Nicolas Rosewick]

Hi,

How can I merge two paired end fastq (R and L) to give a single fastq file ? For information, the sequencing run is 72 bp long and it contains a majority of small RNA (miRNA,...) so a lot of paired end reads will overlap.

For example here's two paired reads :

@HWUSI-EAS529:41:FC62YHFAAXX:8:1:7969:1330 1:N:0:GCCAAT
CTACGAAAGGGCACTTGGAATTCTCGGGTGCCAAGGAACTCCAGTCACGCCAATATCTCGTATGCCGTCTTCTGCT
+
IIIIIIIHIIHIIIIIIHHIIIHGIIIIEIIIIIIEIIHIIIIIIIIIIIHIIIIIBHIHIIHGIGIEGHHEGEEH
@HWUSI-EAS529:41:FC62YHFAAXX:8:1:7969:1330 2:N:0:GCCAAT
AGTGCCCTTTCGTAGGATCGTCGGACTGTAGAACTCTGAACGTGTAGATCTCGGTGGTCGCCGTATCATTAAAAAA
+
IIIIIIIIIIIIIIIIIIIIIIIDHIGIIIHIIIGHGIIIIIIIHHIHIIIIIIIIIHIIIIIIIIHIIGIIIIHI

I find the adapter in the first one:

Code:

EMBOSS_001         1 CTACGAAAGGGCACTTGGAATTCTCGGGTGCCAAGGAACTCCAGTCACGC     50
                                    |||||||||||||||||||||              
EMBOSS_001         1 ---------------TGGAATTCTCGGGTGCCAAGG--------------     21

EMBOSS_001        51 CAATATCTCGTATGCCGTCTTCTGCT     76

EMBOSS_001        22 --------------------------     21

but not in the second one

But I effectively found the overlap between the right read and the left read (using the reverse complement of it)

EMBOSS_001         1 --------------------------------------------------      0

EMBOSS_001         1 TTTTTTAATGATACGGCGACCACCGAGATCTACACGTTCAGAGTTCTACA     50

EMBOSS_001         1 -----------CTACGAAAGGGCACTTGGAATTCTCGGGTGCCAAGGAAC     39
                                |||||||||||||||                        
EMBOSS_001        51 GTCCGACGATCCTACGAAAGGGCACT------------------------     76

EMBOSS_001        40 TCCAGTCACGCCAATATCTCGTATGCCGTCTTCTGCT     76

EMBOSS_001        77 -------------------------------------     76

So my question is, how can I merge the two fastq files to produce a single fastq file?

Thanks,
N.

[manekineko]

Hi, I see that you had a similar case like me, so probably you can help me :)

As I always do the miRNA analysis in single end I'm confused how to proceed when I have paired-end? Can you recommend me how to clean the reads and have them ready for analysis, particularly I cannot understand how and what is the relation of the reverse-compliment miRNA sequence in R2 read to the R1 set?

In summary my R1 read is containing 100nt - miRNA+barcode+smallRNAadapter+another adapter+polyA

my R2 is containing miRNA (reversed compliment to R1) + long adapter (or linker) + polyA

Thanks for any help in advance!

[midox]

Hi,

So for an exact answer to this problem.

The R1.fq are the forward reads and the R2.fq are written in reverse-complement.

For example, if I want to create a single file from reads in R1.fq and R2.fq, I have to do "reverse-complement" of reads in R2.fq?

Am I right?

Thank you

[midox]

no response for this problem?

Answer 1

[pmenzel]

Maybe this program is suited for you: http://www.cbcb.umd.edu/software/flash/

[litiancheng.gansu]

It's no available, the web page cannot be open. http://genomics.jhu.edu/software/FLASH/index.shtml

[matted]

It's here now: http://ccb.jhu.edu/software/FLASH/

Answer 2

[Jeremy Leipzig]

There is a decent program called stitch: https://github.com/audy/stitch

I wrote a script called mergePairs that is very sensitive and incredibly slow: http://code.google.com/p/standardized-velvet-assembly-report/source/browse/trunk/mergePairs.py

Answer 3

[lelle]

As this was referenced from a duplicate question, I will add a newer tool to the list: PANDAseq

Answer 4

[Andreas]

One more: SeqPrep

Andreas

Answer 5

[Charles-Alexandre Roy]

NGmerge (2018) is another option. According to the paper, it performs better than other popular tools like FLASH and PEAR, particularly with respect to the estimation of quality scores for consensus bases.

Answer 6

[Stevelor]

Either use Galaxy or use the single scripts

http://hg.notalon.org/galaxy/galaxy-central/src/7d9bb95caaa7/tools/fastq

HTH!

[Jeremy Leipzig]

uhhh which script?

Answer 7

[Gabriel R.]

If you wish to trim adapters and merge in a single step, you can the leeHom, we use it mainly to reconstruct ancient DNA sequences but it has broader uses as well:

http://nar.oxfordjournals.org/content/42/18/e141

Click here for the Website of the repository

It use a Bayesian maximum a posteriori approach that considers quality scores for both the adapter determination and the merging part.

Answer 8

[FatihSarigol]

There is BBMerge, which is ""designed to merge two overlapping paired reads into a single read. For example, a 2x150bp read pair with an insert size of 270bp would result in a single 270bp read"": http://jgi.doe.gov/data-and-tools/bbtools/bb-tools-user-guide/bbmerge-guide/