Very short contigs and short isotigs of newbler output.

[kentnf]

Hi all,

I am trying to assembly ESTs using newbler 2.5. The minimum length of input ESTs is 100bp and the input ESTs have already been cleaned using seqclean. At the same time, I use the notrim parameter when assembling. But the newbler generates some very short contigs (1bp, 7bp, or 22bp) and short isotigs (55bp). I am very confused about the short contigs and isotigs. I want to know why the short contigs were generated by newbler. Thank you so much.

Here is my assembly parameters:
runAssembly -cdna -urt -notrim -ml 40 -mi 97 input_EST.fasta

[dsenalik]

Lex Nederbragt (flxlex) has a blog post on this very topic:

http://contig.wordpress.com/2011/04/05/newbler-output-iv-on-ultra-short-and-single-read-contigs/

You may want to dig deep deep into the output files to find the connection information between contigs, this is another relevant blog entry

http://contig.wordpress.com/2010/04/13/newbler-output-iii-the-454contiggraph-txt-file/

[kentnf]

Thanks a lot. I checked the 454ReadStatus.txt file. The short contigs is just like exon, one read can be mapped to two contig. That the reason why some contigs are very short, right?

[dsenalik]

Instead of saying like an exon, I would say it's either like two alleles, or like a duplicated copy.