coverage extrapolation failed

[GoogleCodeExporter]

Failed on flow cell C4K6PACXX for the Nicolet samples (NIC125A62-A65)

Other metrics calculated correctly.  Genome says mm10 but it is not, it's human 
(work flow params correctly indicates human).  Part of the problem?

Original issue reported on code.google.com by cmnico...@gmail.com on 8 Jul 2014 at 12:03

[GoogleCodeExporter]

I dont think changing the genome should matter here. 

btw, we had some of cedar bisulfite lanes that were set to mm10, but yaping's 
tool needs mm9, so changed them by hand as well.

anways, in this case, it was because:

TOTAL READS     = 38560343
DISTINCT READS  = 34657396
DISTINCT COUNTS = 29
MAX COUNT       = 43
COUNTS OF 1     = 3.08476e+07
MAX TERMS       = 1000
OBSERVED COUNTS (44)
1   3.08476e+07
2   3.75989e+06
3   23192
4   19884
5   2942
6   1842
7   726
8   501
9   213
10  198
11  94
12  76
13  44
14  40
15  23
16  25
17  16
18  12
19  4
20  8
21  5
22  4
23  6
24  1
26  1
27  1
28  1
35  1
43  1

[BOOTSTRAP ESTIMATES]
________________________________________________________________________________
________________________________________________________________________________
________________________________________
ERROR:  too many iterations, poor sample

so I guess the curve wasn't well defined enough to converge upon a reliable 
projection.  this does not imply good or bad.

Original comment by zack...@gmail.com on 8 Jul 2014 at 4:47

[GoogleCodeExporter]

I do not understand this, there were plenty of reads.  Is it "too complex"
I wonder?

Original comment by cmnico...@gmail.com on 8 Jul 2014 at 4:18

[GoogleCodeExporter]

yeah, i had the same thought. there were so few dups that maybe not enough to 
project.

Original comment by zack...@gmail.com on 8 Jul 2014 at 4:20

[GoogleCodeExporter]

Original comment by zack...@gmail.com on 8 Jul 2014 at 11:34

• Changed state: Done