Is it normal that missing out the spike protein causes no proteins to be detected?

[marklam]

I've been playing around with some data from a PRIDE dataset that has a CHO sample with a spike.
If I analyse the data with just the proteome, I get no proteins, but if I add a fasta with another 3 proteins that were also in the data, I get 11 proteins detected. Is this normal for mass spec searching?

I've made an example with a command-line use of Metamorpheus in this repo: https://github.com/marklam/MetamorpheusData

The default search with just the proteome gives me:

All target PSMs with q-value <= 0.01: 242
All target peptides with q-value <= 0.01: 0
All target protein groups with q-value <= 0.01 (1% FDR): 14
All Precursors: 204336
All MS2 Scans: 122457

TEMP_1_DIA_45 - Target PSMs with q-value <= 0.01: 0
TEMP_1_DIA_45 - Target peptides with q-value <= 0.01: 0

TEMP_1_DIA_45 - Precursors: 68841
TEMP_1_DIA_45 - MS2 Scans: 40868

TEMP_2_DIA_46 - Target PSMs with q-value <= 0.01: 0
TEMP_2_DIA_46 - Target peptides with q-value <= 0.01: 0

TEMP_2_DIA_46 - Precursors: 68081
TEMP_2_DIA_46 - MS2 Scans: 40813

TEMP_3_DIA_47 - Target PSMs with q-value <= 0.01: 0
TEMP_3_DIA_47 - Target peptides with q-value <= 0.01: 0

TEMP_3_DIA_47 - Precursors: 67414
TEMP_3_DIA_47 - MS2 Scans: 40776

The same search with the 3 extra proteins added gives:

All target PSMs with q-value <= 0.01: 3519
All target peptides with q-value <= 0.01: 0
All target protein groups with q-value <= 0.01 (1% FDR): 16
All Precursors: 204336
All MS2 Scans: 122457

TEMP_1_DIA_45 - Target PSMs with q-value <= 0.01: 1260
TEMP_1_DIA_45 - Target peptides with q-value <= 0.01: 0
TEMP_1_DIA_45 - Target protein groups within 1 % FDR: 12
TEMP_1_DIA_45 - Precursors: 68841
TEMP_1_DIA_45 - MS2 Scans: 40868

TEMP_2_DIA_46 - Target PSMs with q-value <= 0.01: 1097
TEMP_2_DIA_46 - Target peptides with q-value <= 0.01: 0
TEMP_2_DIA_46 - Target protein groups within 1 % FDR: 20
TEMP_2_DIA_46 - Precursors: 68081
TEMP_2_DIA_46 - MS2 Scans: 40813

TEMP_3_DIA_47 - Target PSMs with q-value <= 0.01: 896
TEMP_3_DIA_47 - Target peptides with q-value <= 0.01: 0
TEMP_3_DIA_47 - Target protein groups within 1 % FDR: 11
TEMP_3_DIA_47 - Precursors: 67414
TEMP_3_DIA_47 - MS2 Scans: 40776

11 proteins is more than just the ones I added, so I'm confused why this happens.

Thanks for your time!

[trishorts]

too busy to look today. but I will. what kind of mass spec? oribi? ion trap?

[trishorts]

i see. is this dia? cuz we don't really do dia. and it will search but it won't give a good answer. more later. gotta run out the door.

[marklam]

It's Orbitrap DIA data. The results I saw with the 3 added proteins' sequence added looked fine to me (as a relative mass spec novice), but without those it didn't find much.

What is it about DIA data that MetaMorpheus doesn't handle as well? I'd guess that it's the less defined precursor m/z range for the fragmented spectra, but like I say, I'm just learning.

[trishorts]

dia is totally different from DDA. I'm sorry but we don't yet support dia.

[marklam]

I found that the data set does have some DDA samples too, so I tried those out. I've updated the repo I linked at the top with a ResultsDDA folder, but it's still showing that showing no protein groups when just using the proteome, but several protein groups (more than the 3 proteins in the file) when the extras fasta is added:

MetaMorpheus: version 1.1.5
Time to run task: 00:00:35.2782285

--------------------------------------------------

All target PSMs with q-value <= 0.01: 123
All target peptides with q-value <= 0.01: 0
All target protein groups with q-value <= 0.01 (1% FDR): 8
All Precursors: 99207
All MS2 Scans: 61720

control_1_DDA_14 - Target PSMs with q-value <= 0.01: 0
control_1_DDA_14 - Target peptides with q-value <= 0.01: 0

control_1_DDA_14 - Precursors: 32980
control_1_DDA_14 - MS2 Scans: 20698

control_2_DDA_15 - Target PSMs with q-value <= 0.01: 0
control_2_DDA_15 - Target peptides with q-value <= 0.01: 0

control_2_DDA_15 - Precursors: 33157
control_2_DDA_15 - MS2 Scans: 20551

control_3_DDA_16 - Target PSMs with q-value <= 0.01: 0
control_3_DDA_16 - Target peptides with q-value <= 0.01: 0

control_3_DDA_16 - Precursors: 33070
control_3_DDA_16 - MS2 Scans: 20471



--------------------------------------------------

vs

MetaMorpheus: version 1.1.5
Time to run task: 00:00:44.8959726

--------------------------------------------------

All target PSMs with q-value <= 0.01: 4090
All target peptides with q-value <= 0.01: 0
All target protein groups with q-value <= 0.01 (1% FDR): 18
All Precursors: 99207
All MS2 Scans: 61720

control_1_DDA_14 - Target PSMs with q-value <= 0.01: 1406
control_1_DDA_14 - Target peptides with q-value <= 0.01: 0
control_1_DDA_14 - Target protein groups within 1 % FDR: 6
control_1_DDA_14 - Precursors: 32980
control_1_DDA_14 - MS2 Scans: 20698

control_2_DDA_15 - Target PSMs with q-value <= 0.01: 1285
control_2_DDA_15 - Target peptides with q-value <= 0.01: 0
control_2_DDA_15 - Target protein groups within 1 % FDR: 6
control_2_DDA_15 - Precursors: 33157
control_2_DDA_15 - MS2 Scans: 20551

control_3_DDA_16 - Target PSMs with q-value <= 0.01: 1279
control_3_DDA_16 - Target peptides with q-value <= 0.01: 0
control_3_DDA_16 - Target protein groups within 1 % FDR: 5
control_3_DDA_16 - Precursors: 33070
control_3_DDA_16 - MS2 Scans: 20471



--------------------------------------------------

Any ideas what could cause this?

[trishorts]

if you provide me links to the raw files and let me know the organism, i will do a search and let you know what i see. I should be able to determine a strong set of search parameters and explain my choices to you.

[marklam]

Thanks for the help, the files are from this dataset : https://www.ebi.ac.uk/pride/archive/projects/PXD020127
The DDA runs are control_[1/2/3]_DDA_[14/15/16].raw

The organism is Cricetulus Griseus, the file I used is https://raw.githubusercontent.com/marklam/MetamorpheusData/refs/heads/main/Smaller_CHO.fasta which should be equivalent to the UniProt UP000001075 from April 2018, and the fasta for the extra proteins in the sample is https://raw.githubusercontent.com/marklam/MetamorpheusData/refs/heads/main/extras.fasta which should be the Hi3 standard used.