Triqler#

Triqler is a probabilistic graphical model that propagates error information through all steps from MS1 feature to protein level, employing distributions in favor of point estimates, most notably for missing value imputation [THE2018]. The model outputs posterior probabilities for fold changes between treatment groups, highlighting uncertainty rather than hiding it.

quantms & triqler#

quantms exports the triqler input after the quantification steps in the LFQ analysis (Label-free quantification with DDA (LFQ)). The searchScore is computed as 1-PEP. PEP is the Posterior Error Probability. For DIA analysis, the searchScore is computed as 1-Q.Value. The following table is an example how the exported file should looks like:

Table 3 Example of a triqler output generated by quantms#

run

condition

charge

searchScore

intensity

peptide

proteins

6

heart

2

0.9840915

3.275759e07

AAAFEQLQK

O94826

Note

The triqler output is stored in the proteomicslfq folder for label-free (Label-free quantification with DDA (LFQ)) experiments and in diannconvert folder for DIA analysis (Data-independent acquisition (DIA) quantification). Currently Triqler does not support labeled experiments. The triqler output generation automatically activates decoy quantification which makes the pipeline a slower.

Some remarks:

  • For Triqler to work, it also needs decoy PSMs, preferably resulting from a search engine search with a reversed protein sequence database concatenated to the target database. quantms exports the decoy and target proteins into the triqler output.

  • The intensities should not be log transformed, Triqler will do this transformation for you.

  • The search engine scores should be such that higher scores indicate a higher confidence in the PSM. quantms uses a transformation of the Posterior error probability (PEP) as 1-PEP for each PSM.

  • Multiple proteins can be specified at the end of the line, separated by tabs. However, it should be noted that Triqler currently discards shared peptides.

Running Triqler#

Triqler can be run in the quantms output by using the following command:

python -m triqler --fold_change_eval 0.8 out_triqler.tsv

References#

[THE2018]

The M, Käll L. Integrated Identification and Quantification Error Probabilities for Shotgun Proteomics. Mol Cell Proteomics. 2019 Mar;18(3):561-570. doi: 10.1074/mcp.RA118.001018. Epub 2018 Nov 27. PMID: 30482846; PMCID: PMC6398204.