Re: [wgs-assembler-users] Correlation between long unitigs after unitigging and long contigs after scaffolding?

[Walenz, B. <bw...@jc...>]

Larger unitigs almost always lead to better assemblies.  (As one early
developers said: "I've never seen better assemblies from smaller unitigs".)
Unitigs can be split after they are formed (at 1x coverage areas with only
bad mates spanning it) and so the stats out of unitigger aren't exactly what
is input to scaffolder.

Mate happiness (5-consensus-insert-size) probably won't show much difference
here.

We've talked about making scaffolder output size statistics periodically,
but haven't implemented anything.

Even for a running assembly, you can output size statistics for contigs
using tigStore:
 tigStore -g *gkpStore -t *tigStore V -C -d sizes

Where V == the last complete version (has ctg, utg and dat files) in the
tigStore.

A bit heavy weight, but you can (in theory) run terminator using just a
checkpoint and a tigStore, even when scaffolder is running.  Some of the
labeling will be wrong (mate pairs won't be labeled as happy, etc;
contigs/unitigs probably won't be labeled either) but you can get sequence
files.

b



On 8/23/12 5:11 AM, "Ole Kristian Tørresen" <o.k...@bi...> wrote:

> Hi,
> I have several assemblies running, based on different input and
> configurations, and want to have an idea of how well they are doing.
> In the 4-unitigger folder, there is a log2 length histogram. Can I use
> that to get an idea of how well my assembly is going? For example,
> this is from one assembly (bogart):
> checkUnitigMembership()-- 13 (     8192-    16384) 2953
> checkUnitigMembership()-- 14 (    16384-    32768) 168
> checkUnitigMembership()-- 15 (    32768-    65536) 4
> checkUnitigMembership()-- 16 (    65536-   131072) 1
> 
> and this is from another (bog):
> checkUnitigMembership()-- 13 (     8192-    16384) 2302
> checkUnitigMembership()-- 14 (    16384-    32768) 74
> checkUnitigMembership()-- 15 (    32768-    65536) 1
> 
> and a third (bog):
> checkUnitigMembership()-- 13 (     8192-    16384) 2718
> checkUnitigMembership()-- 14 (    16384-    32768) 48
> 
> 
> Since there are more and longer unitigs in the first assembly, will
> that probably turn out to have longer contigs in the end, or is there
> no correlation between this? Is there other places where I can get a
> feel of my assembly? Parsing the scaffold log files in any particular
> way?
> 
> Thank you.
> 
> Ole
> 
> ------------------------------------------------------------------------------
> Live Security Virtual Conference
> Exclusive live event will cover all the ways today's security and
> threat landscape has changed and how IT managers can respond. Discussions
> will include endpoint security, mobile security and the latest in malware
> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
> _______________________________________________
> wgs-assembler-users mailing list
> wgs...@li...
> https://lists.sourceforge.net/lists/listinfo/wgs-assembler-users