query: obtaining SE and PE files from fastq

[ramadatta.88]

Hi all, I am a bit new to assembling stuff. Please spare me and answers even if my questions looks a bit silly.

I obtained an .SRA file ( 200bp PE Illumina reads ) from NCBI traces archive. Using sratoolkit it is converted to fastq format.

(fastq) It looks like this:

@SRR018008.1 307AEAAXX:4:1:1591:659 length=84
GATTTTGAAGGCATATCTTGAAGATGGTGCAGCATCCGAGGTAAGAGACGGGTGAAGCATGGAGCAGAGCGTCAGCAGATGGTG
+SRR018008.1 307AEAAXX:4:1:1591:659 length=84
IIIIIIIIIIIIIIIIIIII?=IIII;IGBI,<:I8:2/5+8IIIIIIIIIIIIIIII:?ID,ID3I1<8C,4:6-.5-+)3(9
@SRR018008.2 307AEAAXX:4:1:1593:693 length=84
GAATTGGCAATCGTCCAATCGTCCAAACGTCCAGAGAGCCGCATAATTACCCGTCACGAATTTCTCTGCTTTGCAGCCGAGCAA
+SRR018008.2 307AEAAXX:4:1:1593:693 length=84
IIIIIIIIIIIIIIIII;II-III0//IH0/;3726,9&5/57IIIIIIIIII,IIIIIBI8A>?)II:AI56;1910056$(%
@SRR018008.3 307AEAAXX:4:1:1672:656 length=84
GGTAGACTTAGTGAATGAACGAAGGGTATCAAAAGATGGAGTCTGCCACCGGCTCGCTGCTCACATCGGAGACCAGCCTGAGCA
+SRR018008.3 307AEAAXX:4:1:1672:656 length=84
IIIIIIIIIIIIIIIIIIEII==IIIH?IIE25-I:G=+5/,IIIIIIIIIIBHI/I490B447-,0-6)90*-%+(('+%*'&


Now my questions are:

1.After conversion SRA-> fastq i obtained only 1 file with read length =84 as shown above. Is it a Single End read (SE) file? If so how can i convert a paired end (PE) file?

2. Is the length=84 present in the file is obtained due to trimming of the ends 100-(8+8) ? if not what does length=84 means?

3. This is bit unrelated to above content:
what is the difference between
a)./velveth sillyDirectory 21 -short data/test_reads.fa
b)./velveth sillyDirectory 21 -short data/test_reads.fa -long data/test_long.fa

Please kindly help me understanding basic concepts. Would be grateful to your answer.

[swbarnes2]

If the data's really single end, you can't convert it to paired end. So you can look at the study that generated the data, and see if they really ran a paired end experiment. If the data's really paired end, but the data's been catted together somehow, you can tell from the names of the reads. You should, for instance, have two different reads with "307AEAAXX:4:1:1591:659" in the name. So grep all your fastqs to see how many times that name pops up. If it only comes up with one read, then it's single end data.

And your reads really are 84 bases long, that's probably why the length says they are 84.

I believe the second velvet command is for the inclusion of Illumina-type data and sanger data together in the assembly of genomes. If you only have Illumina data, that's short reads.

[ramadatta.88]

thank you swbarnes2.

Yeah. I can see two '307AEAAXX:4:1:1591:659' patterns from the file. I can understand it is PE data file.

But, why did the length reduce to 84 in forward ,84 in reverse reads (as shown in previous post), which has been termed 200 bp illumina reads? Is the Read trimmed ? If not Please kindly help me knowing.


Thank you.

[gringer]

The 200bp is probably the total length of each DNA fragment, from which only 84bp of each end have been sequenced:

FFFFFFFFFFFFFFFFFFFFFFFFXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXRRRRRRRRRRRRRRRRRRRRRRRR
  [->forward sequence]    [insert]   [reverse sequence<-]

It's also possible that there's a 200bp insert, rather than 200bp fragment length.