understanding dwgsim_eval output

[oiiio]

Here is the command line:

dwgsim_eval -b -g 0 1300000.bwa.01.sorted.bam

Here is the first line of the output:

60           13 / 2111834                 13       2111834       2600004  6.156e-06  6.156e-06  5.000e-06  5.000e-06

There were 1300002 pairs so a total of 2600004. I read what information was on the dwgsim website, but I am still having trouble understanding how to read this output. Any help/comments/input at all would be greatly appreciated.

[nilshomer]

I would be happy to explain given specific questions, and then update the documentation: http://sourceforge.net/apps/mediawik..._-_dwgsim_eval

[oiiio]

Thanks, and for starters does the 13 / 2111834 indicate that 13 out of 2111834 reads mapped correctly at that threshold? The second column ('mc' on the documentation) indicates this, correct?

[nilshomer]

Could you print the line in question?

[oiiio]

60           13 / 2111834                 13       2111834       2600004  6.156e-06  6.156e-06  5.000e-06  5.000e-06

Here is the line in question from my output.
The only thing that I understand for sure is that 60 in the first column is the minimum mapping quality threshold. The documentation indicates 19 columns in the output, whereas I have only 9. Maybe some columns are showing null values?

This might be asking a bit much, but could you tell me the meanings of '13/2111834' , '13' , '2111834', and '2600004'? Perhaps that could help me understand what is happening here.

Again, i used parameters '-g 0' and '-b' for BWA.

[nilshomer]

I think you have an old version. I released a dwgsim package update yesterday.

Nils

[oiiio]

In function "process_bam": Warning[OutOfRange]. Variable/Value: 2L 17664817 17664105 1 0 4 0 0 2 0 0 116ac.
Message: [dwgsim_eval] read was not generated by dwgsim?.

Does the new version of dwgsim_eval require that reads were generated by the newer version of dwgsim ?

[nilshomer]

Yes it does unfortunately. Those two are intimately paired. Also, you can subscribe to [email protected] to get notified with updates.

[oiiio]

Are there any errors in the old versions of dwgsim and _eval? I wanted to verify results from my own BAM parser that I am using on dwgsim data.

If possible do you think you could help me understand what I am seeing in my output from the old version?

[nilshomer]

Sure, could you give me the version #? You might be able to get it when you run dwgsim without any options.

[oiiio]

The version is 0.1.2

[nilshomer]

Caveat: I really like the new version a lot better, and there may be bugs in the old version. Below are the columns:
#1: metric
#2: # mapped correctly at the threshold
#3: # of mapped at the threshold
#4: # mapped correctly at or greater than the threshold
#5: # of mapped at or greater than the threshold
#6: # of reads total
#7: fraction of reads mapped correctly of those mapped, at the threshold
#8: fraction of reads mapped correctly of those mapped, at or greater than the threshold
#9: fraction of reads mapped correctly out of all reads, at the threshold
#10: fraction of reads mapped correctly out of all reads, at or greater than the threshold

[oiiio]

Awesome, thanks. I will upgrade to the newer version, but for now I have a ton of data that was generated with my older one.