compare_peaks: compare_peaks

In YasinEl/mzRAPP: Benchmark Dataset Generation and Non-Targeted Data Pre-Processing Assessment

compare_peaks

Description

Matches peaks reported in the aligned and unaligned non-targeted output against the provided benchmark as prepared by check_benchmark_input. For details on how this matching procedure is conducted please check the mzRAPP readme https://github.com/YasinEl/mzRAPP#matching-between-bm-and-npp-output-background.

Usage

compare_peaks(b_table, ug_table, g_table, algo)

Arguments

b_table	output from check_benchmark_input
ug_table	one of the listed objects (ug_table) in output of check_nonTargeted_input. e.g. check_nonTargeted_input_output$ug_table
g_table	one of the listed objects (g_table) in output of check_nonTargeted_input. e.g. check_nonTargeted_input_output$g_table
algo	output format of ug_table and g_table. Can be "XCMS", "XCMS3", "Metaboanalyst", "SLAW", "El-Maven", "OpenMS", "MS-DIAL", "CompoundDiscoverer" or "MZmine 2"

Details

The output of this function is a list containing 9 different elements. In the following benchmark peaks are referred to as BM, non-targeted unaligned peaks as NP and NP which have been aligned across samples (features) as NF. In all following tables (provided as data.table objects) benchmark variables are labeled with a "_b", unaligned non-targeted peaks via "_ug" and aligned non-targeted peaks via "_g" suffix.

BM_NPPoutput_size: Gives peak and aligned feature counts for the used benchmark as well as non-targeted outputs

nr_of_b_peaks: benchmark peak count

nr_of_b_features: benchmark feature count

nr_of_ug_peaks: NP count

nr_of_g_peaks: number of peaks reported in the aligned output

nr_of_g_features: NF count

algorithm: non-targeted output format

Overview_per_molecule: Provides overview of different performance metrics per benchmark molecule. (metrics are summed up over whole molecule)

molecule_b: molecule name

Min.er: count of alignment errors as defined in the mzRAPP readme

BM.div: count of alignment divergences as defined in the mzRAPP readme

lost: count of NP for which no match among NF was found (NP lost during alignment)

R_pp: count of random (high) missing values in the unaligned non-targeted output as defined in mzRAPP readme

S_pp: count of systematic (low) missing values in the unaligned non-targeted output as defined in mzRAPP readme

R_ft: count of random (high) missing values in the aligned non-targeted output as defined in mzRAPP readme

S_ft: count of systematic (low) missing values in the aligned non-targeted output as defined in mzRAPP readme

IRb_ok_pp: count of isotopologue ratios calculated from NP classified as "good" as defined in the mzRAPP readme

IRb_off_pp: count of isotopologue ratios calculated from NP classified as "bad" as defined in the mzRAPP readme

IRb_ok_ft: count of isotopologue ratios calculated from NF classified as "good" as defined in the mzRAPP readme

IRb_off_ft: count of isotopologue ratios calculated from NF classified as "bad" as defined in the mzRAPP readme

Split_peaks: count of splitted peaks as defined in the mzRAPP readme

Found_peaks_ft: count of BP for which a match among NF was detected. For details on the matching procedure see the mzRAPP readme

Not_Found_peaks_ft: count of BP for which no match among NF was detected. For details on matching procedure see the mzRAPP readme

Found_peaks_pp: count of BP for which a match among NP was detected. For details on matching procedure see the mzRAPP readme

Not_Found_peaks_pp: count of BP for which no match among NP was detected. For details on matching procedure see the mzRAPP readme

Extra_feature_matches_ft: Number of additional NF matching on a given benchmark feature on top of best match

Extra_peak_matches_pp: Number of additional NP matching on a given BP on top of the best match

Matches_BM_NPPpeaks: Extensive table allowing to inspect matches between the benchmark and the unaligned non-targeted output in detail. Unaligned non-targeted peaks are also matched to aligned non-targeted peaks via reported abundance values in order to count alignment errors. In order to allow the inspection of all input variables all variables available in benchmark and the non-targeted output are kept. It is worth noting that more than one match can occur per BP. For the best match (as defined in the mzRAPP readme) the variable main_peak equals TRUE.

Unmatched_BM_NPPpeaks: Benchmark peaks for which no match could be found.

SplittedMatches_BM_NPPpeaks: All splitted matches (as defined in the mzRAPP readme) are summarized in this table. Splitted matches for which also a full match exists in in the Matches_BM_NPPpeaks table have "present_in_found" = TRUE

MissingPeak_classification: Classifies the missing value status of peaks before and after alignment. This classification is made only correct alignment of the benchmark is in general agreement with the alignment of the non-targeted output per molecule (as described in mzRAPP readme). Peaks for which this was confirmed show "Connected" = TRUE. The MissingPeak status is determined for the unaligned ("missing_peaks_ug") as well as the aligned ("missing_peaks_g") output individually and can hold the following values.

F (found): Peak is not missing

NC (not confirmable): alignment could not be confirmed as described above

L (lost): all peaks of the respective features were not found

R (random): random (high) missing value as defined in mzRAPP readme

S (systematic): systematic (low) missing value as defined in mzRAPP readme

IT_ratio_biases: Table showing the status of calculated isotopologue ratio biases as compared to enviPat predicted ratios calculated from abundances reported from BP ("benchmark"), NP ("NPP_peak picking") and NF ("NPP_features"). Biases are classified for NP ("diffH20PP_pp") and NF ("diffH20PP_ft") into good ("Inc. < 20 percentage points") or bad ("Inc. > 20 percentage points") as defined in the mzRAPP readme.

AlignmentErrors_per_moleculeAndAdduct: Overview of alignment problems per molecule and adduct.

Min.errors: count of alignment errors as defined in the mzRAPP readme

BM.div: count of alignment divergences as defined in the mzRAPP readme

Lost_b.A: count of NP for which no match among NaP was found (NP lost during alignment)

Matches_BM_NPPpeaks_NPPfeatures: Table containing matched non-targeted unaligned and aligned peaks for each BP with their respective reported abundances.

min_mz_start: lowest mz allowed for matched NF to be considered (as described in mzRAPP readme)

max_mz_end: highest mz allowed for matched NF to be considered (as described in mzRAPP readme)

min_rt_start: lowest RT allowed for matched NF to be considered (as described in mzRAPP readme)

max_rt_end: highest RT allowed for matched NF to be considered (as described in mzRAPP readme)

peak_area_b: peak area as reported in benchmark

peak_area_ug: peak area as reported in unaligned NP

area_g: peak area as reported in NF

Value

returns list containing different tables including data from different types of comparisons

YasinEl/mzRAPP documentation built on Feb. 18, 2024, 11:49 a.m.