Paired end reverse orientation?

[darthsequencer]

Hi,
I was given a lot of Illumina paired end reads. ~300bp inserts were used.

If I wanted to combine these into one long read where it would be:

Read 1 ----- Read 2

Where --- = the bases between the ends.

Do I just need to reverse Read 2? or reverse complement?

I think reverse and not complement.

Thanks

[mastal]

Paired end

I think you need to reverse complement.

My understanding is that the second read comes from the opposite strand to the first read.

Maria

[darthsequencer]

Ok thanks - if anyone else can confirm that'd be really helpful.

Thanks

[BAMseek]

Yep, the second read of an Illumina paired-end read would need to be reversed and complemented. Don't forget to reverse the quality string, too. You might want to check out FLASH, which will detect paired end reads that overlap each other and merge them into a single read.

Justin

[darthsequencer]

Thanks for the help guys.

[modi2020]

Hi guys...
I am a bit confused about the way we deal with paired end reads...
Say we have a read that is AAACGCGCTTTCCC
does paired end mean we only sequence AAA and CCC?
how about the sequence in between?

[BAMseek]

Hi guys...
Say we have a read that is AAACGCGCTTTCCC
does paired end mean we only sequence AAA and CCC?
how about the sequence in between?

In this case, AAACGCGCTTTCCC would be the sequence fragment. Ideally, you would just read the entire fragment, but due to the current limitations of the technology, that may not be possible. Paired-end sequencing allows you to read some bases from the beginning and some bases from the end of the fragment. If the fragment is short enough, the first read might overlap the second read. But in cases where the fragment length is longer than the sum of the paired read lengths, you would have a gap in the middle.

Let's say your read length was 4 base pairs. Then the first read would be AAAC and the second read would be GGGA (because the second read is obtained by reading the reverse complement).

That's at least my understanding of the process. Hopefully that is of some help.

Justin

[modi2020]

Thank you BAMseek for helping me with this.
I have an inquiry though. Assuming the reads are 200 bp. suppose we read the ends only. How would we know the sequence in between?
Also, while sequencing (by synthesis) wouldn't we get a TTTG read for the AAAC sequence ?
I thought that in sequencing by synthesis we add the compliment of the base in the genomic sequence. Thank you

In this case, AAACGCGCTTTCCC would be the sequence fragment. Ideally, you would just read the entire fragment, but due to the current limitations of the technology, that may not be possible. Paired-end sequencing allows you to read some bases from the beginning and some bases from the end of the fragment. If the fragment is short enough, the first read might overlap the second read. But in cases where the fragment length is longer than the sum of the paired read lengths, you would have a gap in the middle.

Let's say your read length was 4 base pairs. Then the first read would be AAAC and the second read would be GGGA (because the second read is obtained by reading the reverse complement).

That's at least my understanding of the process. Hopefully that is of some help.

Justin

[BAMseek]

Assuming the reads are 200 bp. suppose we read the ends only. How would we know the sequence in between?

You would have to rely on other reads that align nearby to get the bases in between. Your fragmentation/sonification size selection would determine how big of a gap you expect between read pairs.

Also, while sequencing (by synthesis) wouldn't we get a TTTG read for the AAAC sequence ?
I thought that in sequencing by synthesis we add the compliment of the base in the genomic sequence.

Yes, that's correct. But also keep in mind that you might be reading the cDNA of the original sequence, and the sequence might have gone though a few rounds of PCR. If the protocol is not strand specific, then the strand information is lost anyway. I think the important thing to note that if it is Illumina sequencing, you would expect the strands of a read pair to be opposite of each other (forward-reverse or reverse-forward) in a "proper" pair.

[tahamasoodi]

Hi BAMseek,
I also have a doubt in this. If we read the ends (say 100 bp) only for a fragment of 500 bp, there will be around 300 bp in between which would not be read. You say we have rely on other reads that align nearby, which are these reads if we have only forward and reverse reads for a fragment?

[kwaraska]

The basis of this method of sequencing is that you have random shearing.

While one fragment will read bases 1-300, the next one might read 200-500, and a third may be 100-400. In this way, every base gets read, but on different fragments.

[tahamasoodi]

Thanks Kwaraska, it was really helpful!