search

Nesvilab / IonQuant

A label free quantification tool.
Other
15 stars 8 forks source link

MBR halting error #23

Closed m-a-valach closed 3 years ago

m-a-valach commented 3 years ago

Having updated philosopher to v4.0.0, I also updated IonQuant to v1.7.5 as recommended — and all ran smoothly when not using the MBR option. However, when it is on (mbr=1), IonQuant comes to a halt once it has finished calculating the correlations between the runs: 2021-08-11 17:29:08 [INFO] - With ion FDR 0.010000, ion probability threshold is 0.981200 2021-08-11 17:29:08 [INFO] - With peptide FDR 1.000000, peptide probability threshold is -0.000100 2021-08-11 17:29:08 [INFO] - With protein FDR 1.000000, protein probability threshold is -0.000100 2021-08-11 17:29:08 [INFO] - Updating Philosopher's tables... java.lang.NullPointerException at q.<init>(Unknown Source) at ionquant.IonQuant.main(Unknown Source) 2021-08-11 17:29:09 [ERROR] - null

fcyu commented 3 years ago

Looks like the same issue as https://github.com/Nesvilab/FragPipe/issues/420. Can you help to check if there are empty Prev AA or Next AA in your psm.tsv.

Can you also check if you let Philosopher print decoys?

If not, can you send us all of your files?

Thanks,

Fengchao

m-a-valach commented 3 years ago

Right, for decoys, Prev AA and Next AA in psm.tsv are empty. Printing of decoys in Individual reports is set to true. So, setting it to false should resolve the issue?

fcyu commented 3 years ago

Yes, setting it to false should be able to resolve this issue. Also, there is no need to print decoys for most cases because the PSMs already pass the FDR filtering.

Best,

Fengchao

m-a-valach commented 3 years ago

Thanks, Fengchao. The run completed without any hiccups after the suggested modification (in Philosopher's pipeline configuration file).