import of Spectronaut 9 report into MSstats fails

Tobias Kockmann

unread,

Jul 19, 2016, 3:56:11 AM7/19/16

to MSstats

Dear MSstats,

I recently started using Spectronaut 9 to process DIA data. Now, I would like to continue the statistical analysis in MSstats, but the dataProcess() function terminates with an error due to duplicated features. Here is a screenshot of the corresponding log file (I also attached the complete file):

There is two things happening: (i) Spectronaut does not export features with censored intensities (NAs), therefor the report is incomplete. MSstats corrects this by filling up the missing lines. Interestingly, it afterwards complains about the same features being duplicated and terminates the process. Has anyone every experienced this phenomenom?

Here is some information about the df used as input for dataProcess():

> str(MSstats.report)

'data.frame': 1715433 obs. of 10 variables:

$ Condition : Factor w/ 3 levels "ko","l","nl": 1 1 1 1 1 1 1 1 1 1 ...

$ Run : Factor w/ 15 levels "20160609_06_classicDIA_37757.htrms",..: 1 1 1 1 1 1 1 1 1 1 ...

$ BioReplicate : Factor w/ 15 levels "20160609_06_classicDIA_37757.htrms",..: 1 1 1 1 1 1 1 1 1 1 ...

$ ProteinName : Factor w/ 2516 levels "A0AVT1","A0FGR8",..: 1 1 1 1 1 1 1 1 1 1 ...

$ PeptideSequence : Factor w/ 19585 levels "_[-131]MAALRPLVKPK_",..: 13085 13085 13085 13085 13085 13085 12547 12547 12547 12547 ...

$ PrecursorCharge : int 2 2 2 2 2 2 2 2 2 2 ...

$ IsotopeLabelType: Factor w/ 1 level "L": 1 1 1 1 1 1 1 1 1 1 ...

$ Intensity : num 33977 33977 33977 33977 33977 ...

$ FragmentIon : Factor w/ 125 levels "a10","a11","a12",..: 125 84 63 45 124 85 121 122 115 104 ...

$ ProductCharge : int 1 1 1 1 1 1 1 1 1 1 ...

> head(MSstats.report)

Condition Run BioReplicate ProteinName PeptideSequence PrecursorCharge IsotopeLabelType Intensity

1 ko 20160609_06_classicDIA_37757.htrms 20160609_06_classicDIA_37757.htrms A0AVT1 _QDVIITALDNVEAR_ 2 L 33976.62

2 ko 20160609_06_classicDIA_37757.htrms 20160609_06_classicDIA_37757.htrms A0AVT1 _QDVIITALDNVEAR_ 2 L 33976.62

3 ko 20160609_06_classicDIA_37757.htrms 20160609_06_classicDIA_37757.htrms A0AVT1 _QDVIITALDNVEAR_ 2 L 33976.62

4 ko 20160609_06_classicDIA_37757.htrms 20160609_06_classicDIA_37757.htrms A0AVT1 _QDVIITALDNVEAR_ 2 L 33976.62

5 ko 20160609_06_classicDIA_37757.htrms 20160609_06_classicDIA_37757.htrms A0AVT1 _QDVIITALDNVEAR_ 2 L 33976.62

6 ko 20160609_06_classicDIA_37757.htrms 20160609_06_classicDIA_37757.htrms A0AVT1 _QDVIITALDNVEAR_ 2 L 33976.62

FragmentIon ProductCharge

1 y9 1

2 y10 1

3 b3 1

4 b12 1

5 y8 1

6 y11 1

Greetings,

Tobi

msstats-3.log.zip

Tobias Kockmann

unread,

Jul 20, 2016, 7:35:00 AM7/20/16

to MSstats

Dear MSstats support,

I meanwhile figured out where the problem was. The export scheme generated some duplicated IDs and I was able to fix this issue. Now, the dataProcess() function runs as expected and reports the right number of features and replicates in the different groups...but, the summarisation of feature intensities by Tukey's median polish fails or to be more precise it does not finish. I have been running MSstats for several hours now and there is no progress in status messages coming from dataProcess(). Interestingly, the protein was processed last is one with a high number of matched peptides (several hundred). Could it be that the median polish is not working robustly which such high numbers of features?

Greetings,

Tobi

Meena Choi

unread,

Jul 20, 2016, 9:51:53 AM7/20/16

to MSstats

The large number of features in a protein does not issue for TMP. I guess that that issue came from imputation procedure and there are some other issues about missing patterns or something else in the proteins which can not be fitted. But, It is hard to say why without seeing the data. If you can share the subset of data for that specific protein, we can investigate it.

- Meena

Tobias Kockmann

unread,

Jul 25, 2016, 6:33:26 AM7/25/16

to MSstats

Hi Meena,

I could get rid of the error by using the topN option in the dataProcess() function.