PIA Analysis

This is the PIA Analysis for a PIA intermediate file generated by the PIA Compiler.

Options

general settings

column to PIA XML file (on port 0)
Select the column of the incoming table, which holds the PIA XML file created by the PIA Compiler. Only the first entry will be used. If either input port 1 or 2 are used, don't set this.
Fail on no decoys
Select this if a decoy database was used for spectrum identification. This checks, whether any decoy was detected using the settings on the PSM level and stops the execution, if none was found, which is probably due to erroneous settings.
Create PSM sets
Select this option, if multiple search engines were used for spectrum identification. PIA will try to put identifications of the same spectrum in one set and allows the calculation of the Combined FDR Score as a meta score combining all search engines results in one PSM set.
consider modifications to distinguish peptides
Usually a peptide is defined only by the amino acid sequence. Select this, to distinguish peptides also by modifications.
Export level
Currently only one file can be exported and will be written to output port 3. Select here, which level (PSMs, peptides, proteins) should be exported or none to disable export.
Export format
Select the format of the exported file.

PSMs settings

FileID for PSM output
PIA exports the PSMs of the merge/overview of all files or one file only at the moment. Select 0 for the overview/merge or 1..n for a specific file. You can get the IDs of the compiled files by inspecting the summary view of the compiler. Usually, the order of the files given to the compiler is maintained.
Calculate FDR for all files
Select whether the FDR for all files should be estimated, based on the Target Decoy Approach. This includes calculation of the FDR Score (a smoothed FDR), on PSM and other supported levels. This is needed for merging the results of multiple search engines with the Combined FDR Score.
Calculate Combined FDR Score
The Combined FDR Score can be used to combine the results of multiple search engine runs on a single file. Needs creation of PSM sets on general settings.
accession pattern
Select this option if the spectra were identified against a target-decoy-database and decoys entries can be identified by a regular accession.
by searchengine
Select this option if entries were assigned as target or decoy by the search engine (e.g. when selecting "Decoy search" in Mascot) or a preprocessing step.
only top identification
Use only the best identification per spectrum and discard all other identifications (they will get NA as FDR and can thus be filtered).
all identifications
Use all identifications per spectrum, not only the top scoring. Sometimes the more probable identification does not get the best rank.
Preferred PSM scores
Select the preferred scores used for the FDR estimation. If a file contains one of these scores, it will be used for the FDR estimation of this file. If not, or no score is selected, default scores will be used.
Filters for PSM level
Selected filters here will be used for the creation of the PSM table on port 0 and for PSM level export. It will not interfere with peptide or protein inference.

peptides settings

Infere peptides
Protein inference can be turned off, does not interfere with protein inference. Can save some time, if peptide level information is not needed.
FileID for peptide output
PIA exports the peptides of the merge/overview of all files or one file only at the moment. Select 0 for the overview/merge or 1..n for a specific file. You can get the IDs of the compiled files by inspecting the summary view of the compiler. Usually, the order of the files given to the compiler is maintained.
Filters for peptide level
Selected filters here will be used for the creation of the peptide table on port 1 and for peptide level export. It will not interfere with protein inference.

proteins settings

Infere proteins
Turn off protein inference, if no protein level data should be generated.
Inference method
Select the inference method here. Spectrum Extractor is the recommended method, while "Report all" is no inference, but just reports all possible protein groups from the PSM data.
Inference filters
Select filters here for the PSMs and peptides used for the protein inference. Here e.g. the FDR filtering on PSM level should be selected.
Scoring method
Select "additive scoring" for scores, which have a higher score better (e.g. Mascot Ion Score, X!Tandem Hyperscore), or "multiplicative scoring" for scores with smaller score better (e.g. FDR Score, Mascot Expect). Use the geometric mean scoring only, if you know, what you are doing.
Basescore for protein scoring
Select the PSM score used for the calculation of the protein scores here. If the (Combined) FDR Score was calculated, it is recommended to use it. Take care to select a score, that is created by PIA or the search engine(s).
PSMs used for scoring
By default, only the best PSM score per peptide is used for protein scoring. If multiple PSMs per peptide should influence the score, use all PSMs.
Filters for protein level
Selected filters here will be used for the creation of the protein table on port 2 and for protein level export. It will not directly interfere with protein inference, but only the representation of the inferred groups in the mentioned exports.

Privacy information

Collection of anonymous usage statistics
The PIA team is collecting usage statistics for quality control and, most important, funding purposes. If you don't like to provide the anonymous usage data, disable it in the KNIME preferences.

Input Ports

Icon
A PIA XML file from a table, may be gzipped. This would be the default output of the PIA Compiler. Either this or a file in port 1 must be given.
Icon
A PIA XML file from a file port, e.g. from Input File. Either this or a file in port 0 must be given. Only the first suitable file in the port will be used.
Icon
An optional spectrum file for the spectrum viewer (not yet fully implemented).

Output Ports

Icon
The results on the PSM level.
Icon
The results on the peptide level.
Icon
The results on the protein level.
Icon
The generated export file (not yet implemented for CSV).

Views

PIA Result Analysis
The PIA Analysis view: browse the PSM, peptide and protein results and visualise the results.
PSM Spectrum viewer
Spectrum viewer on PSM level. Node must be connected to a spectrum file to create useful output.

Workflows

Links

Developers

You want to see the source code for this node? Click the following button and we’ll use our super-powers to find it for you.