how to make SRA downloaded fastq (PacBio RS I reads) acceptable for Falcon

[jw457]

I am new in Pac-Bio seq analysis and doing some initial test run in Falcon.

I have a downloaded dataset (Pac-Bio RSI reads) from SRA, and converted it to *.fastq, it looks like this:

@SRR1168519.1 length=302
ATTTTTGTCTGTCCGATTCTGATAGCAGGC
GCATATCAGATGAATCTGATGAGTCAACACTGGTTGGTTCGTTGCTCAGTAGTTATGTTCGTGTGGAGCGTCGTATTGGTATCGAGTCTGATTGTCAGTCATCGATGGTCATTAGTCACGTCCTTCCAGTAGTTCGTATCAACATGCTTCACTATTCTTGTTGTTGTAGATGTTATTCGTATTAGTGTGAGTGTCAGTAGTTACGCGTACAGTATCGGGATTTCGTAGCAGCGCGCGGCGTTGCGGAGTCAAGATTCATGGCTGGACTACGG
+SRR1168519.1 length=302
!"!!!"#$"##!!!"!!"!"#""""#$#"!"!""!!!!!""%"""!"!"#""!#"!!!"!#"!#!!!"!!!"""!!!!"""#!!"#"!"!""!"!!!!""#!!!""!!!"!#!"###"#""!"!!!##!#!#!"!"""!"$$!!"#"$""#"!!"!!#"!!#!!!"!"""!!""%#"$#"$"#"!!!"!!!!!"!!!"!"!"!$#%&%%$"""""""!#"!"!!""##"$!!!!!!!$$!!!!!#!!"!!!!%!"$"!!"""!!!!!!!"!!!!!!$$#"!"!!!"!$$#"!$!!!""!"""

Even after using Falcon-formatter for fasta format conversion, it does NOT work in Falcon.

And I know that the fasta files require strict formatting with the information of movie, time of run start, SMRT barcode, etc. and should look like this:
>m140913_050931_42139_c100713652400000001823152404301535_s1_p0/9/1607_26058 RQ=0.831
TGGCATCTCATAAAGCCGCGCGGACGGGCAATAGCACTGGTTCGATTGTCTGGTGTTTATTCCCGGCTGT
TGGGCTGAGTTTGTGATCCCGGTGAACTTCTCGCATGCCGACAGCATCATGATCGGTGCGCTGTCTCCCT
GGCAAATAGAAGTTGTTCAATAACGCGCGCGACTGGCCGTTGGCCTCGGGCGGTTAGCGATGCATCGATG
TTTGCTGGGCTGCTAATTGTGCCCGATAATATGGTTGGTTCGGCACTAAACGACCAGCAAAAAAAAGCGT
GGGAGAACAGATGAAATTATTTACGCGGTAGTTCGTTTCGCCGCTGGCGGATTGTGATTTTGCTGGCTTG
GTCTTACCGTTTTCCTCTACGCGGCCCAATGCTGAGCTGGGTATCTATTCGTTATACGGCTCTGAAGGCT

My question:

1. Can I make up some dummy variables equivalent for ">m140913_050931_42139_c100713652400000001823152404301535_s1_p0/9/1607_26058" to make Falcon work properly?

2. Or is there another way to recover *.sra file I downloaded to make it work properly in Falcon? I downloaded *.sra and then do fastq-dump.
Does this process lead to the loss of the head information such as "m140913_050931_42139_c100713652400000001823152404301535_s1_p0/9"?

Looking forward to your help!

Many thanks!

[flxlex]

I was going to point you to http://www.ebi.ac.uk/ena/data/view/SRR1168519 and suggest downloading the 'Submitted files (ftp)', but there are none. I have two alternative solutions left:

- try to contact the original submitters and ask them for the file(s) they uploaded
- reverse engineer the names with some fake information that makes the IDs it compatible with FALCON (this requires some scripting)

Finally, I don't think you are the only one with this problem. I was going to suggest you file an issue on the FALON github pages, but I see you already have https://github.com/PacificBiosciences/FALCON/issues/428 :-)

[GenoMax]

If you click on the Download tab for these runs (example in link) then the "bas.h5" files as submitted are available. If you have access to SMRTportal you may be able to do something with them (though you may still need to create the metadata files).

[flxlex]

Nice, @GenoMax, I didn't know about that. Now if only ENA would add that link...

[colindaven]

I got quite a long way with this script:

https://github.com/PacificBiosciences/FALCON/issues/251

Also had to remove duplicates