Seeming error in delta_cn in Comet output via crux

[Erik Johnson]

Hello,

Thanks in advance for your help.

I'm running Comet via the crux tool. I'm seeing delta_cn values that don't make sense to me. E.g., in the attached output file, the top PSM has xcorr score = 4.2550 and the next best PSM has xcorr score = 3.2040 so I'd expect the delta_cn to be (4.2550-3.2040)/4.2550 = 0.2470 but the delta_cn in the file is 0.2844 != 0.2470. 

So it seems like there's a bug or I'm missing something. 

Thanks for your hlep!

Best,

Erik

[Michael Riffle]

By default, comet calculates this by finding the first "dissimilar peptide" ID for that scan. I don't know the exact logic it uses, but I'm guessing in this case it's using the 4th hit starting with "EVE" as the first dissimilar hit, which gives (4.255-3.041)/4.255 = 0.2853

This still isn't exactly the same as the reported delta cn, but could be caused by rounding the reported xcorr scores.

Mike

[Jimmy Eng]

Mike is correct.  Comet by default will calculate the delta_Cn between the top hit and first first dissimilar peptide.  The logic to determine what a similar peptide is can be found in CometPostAnalysis::CalculateDeltaCn().  The similarity threshold is 75% sequence identity based on the very poor peptide similarity logic implemented in the code. 

If you do want to have Comet report the delta_Cn between the top two hits and not use the similarity logic, there is a parameter that allows you to control this:  explicit_deltacn.  This is specified in Crux Comet using the "--explicit_deltacn" command line option.

Jimmy

--
You received this message because you are subscribed to the Google Groups "crux-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to crux-users+...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/crux-users/95f4374f-177d-4ba7-82bb-b5793fb1a7aan%40googlegroups.com.

[Erik Johnson]

Thanks Jimmy and Mike! Makes sense. I've got another question that I'll open another thread for so that the threads are separated. 'preciate it!