search

loneknightpy / idba

124 stars 53 forks source link

Detailed Docs? #8

Closed sklages closed 6 years ago

sklages commented 8 years ago

Hi,

is there any documentation available apart from <idba_hyb/ud> without parameters? E.g. how do the '--read_level_n' parameter work? If i had two MP lib, I'd use '--read_level_2' for the first and '--read_level_3' for the second lib? Is the MP insert size relevant for these parameters?

The command output in general is really minimal; a more detailed description of the parameters would be really great.

best, Sven

jvollme commented 8 years ago

Just to keep this question in discussion: Concerning the read_levels: Based on the principle of other assemblers such as SOAP, I would assume that it's usually best to sort different libraries by insert size and to use the smalles insert size first for intital scaffolding. And then to add the next highest insert-size library for scaffolding the initial scaffolds, and so on... Therefore you would add the smallest insert size library as "-r", the second-smallest as "--read_level2" and so on (I assume). Mate-pairs would therefore be added as one of the highest read-levels (again I assume). IDBA_UD seems to support upto 5 different library "channels" so if you would have more than that it would probably make sense to merge datasets from libraries which share a similar insert size.

Are these assumptions correct?

loneknightpy commented 8 years ago

@jvollme You are right. The read libraries are applied sequentially, read, --read_level2, etc.