, Dinesh Barupal [cre,
aut] | Title: | United Formula Annotation (UFA) for HRMS Data Processing |
|---|---|
| Description: | A pipeline to annotate chromatography peaks from the 'IDSL.IPA' workflow <doi:10.1021/acs.jproteome.2c00120> with molecular formulas of a prioritized chemical space using an isotopic profile matching approach. The 'IDSL.UFA' workflow only requires mass spectrometry level 1 (MS1) data for formula annotation. The 'IDSL.UFA' methods was described in <doi:10.1021/acs.analchem.2c00563> . |
| Authors: | Sadjad Fakouri-Baygi [aut]
, Dinesh Barupal [cre,
aut] |
| Maintainer: | Dinesh Barupal <[email protected]> |
| License: | MIT + file LICENSE |
| Version: | 2.0 |
| Built: | 2026-05-19 08:20:30 UTC |
| Source: | https://github.com/idslme/idsl.ufa |
aggregatedIPdbListGenerator
aggregatedIPdbListGenerator(MassMAIso)aggregatedIPdbListGenerator(MassMAIso)
MassMAIso |
MassMAIso |
AggregatedList
This function detects frequent molecular formulas across multiple samples on the aligned peak table matrix.
aligned_molecular_formula_annotator(PARAM)aligned_molecular_formula_annotator(PARAM)
PARAM |
a parameter driven from the UFA_xlsxAnalyzer module. |
This function sorts a vector of molecular formulas to aggregate organic compound classes with repeated/non-repeated substructure units. This function only works for molecular formulas with following elements: c("As", "Br", "Cl", "Na", "Se", "Si", "B", "C", "F", "H", "I", "K", "N", "O", "P", "S")
detect_formula_sets(molecular_formulas, ratio_delta_HBrClFI_C = 2, mixed.HBrClFI.allowed = FALSE, min_molecular_formula_class = 2, max_number_formula_class = 100, number_processing_threads = 1)detect_formula_sets(molecular_formulas, ratio_delta_HBrClFI_C = 2, mixed.HBrClFI.allowed = FALSE, min_molecular_formula_class = 2, max_number_formula_class = 100, number_processing_threads = 1)
molecular_formulas |
a vector of molecular formulas |
ratio_delta_HBrClFI_C |
c(2, 1/2, 0). 2 to detect structures with linear carbon chains such as PFAS, lipids, chlorinated paraffins, etc. 1/2 to detect structures with cyclic chains such as PAHs. 0 to detect molecular formulas with a fixed structures but changing H/Br/Cl/F/I atoms similar to PCBs, PBDEs, etc. |
mixed.HBrClFI.allowed |
mixed.HBrClFI.allowed = c(TRUE, FALSE). Select 'FALSE' to detect halogenated-saturated compounds similar to PFOS or select 'TRUE' to detect mixed halogenated compounds with hydrogen. |
min_molecular_formula_class |
minimum number of molecular formulas in each class. This number should be greater than or equal to 2. |
max_number_formula_class |
maximum number of molecular formulas in each class |
number_processing_threads |
Number of processing threads for multi-threaded computations. |
A matrix of clustered classes of organic molecular formulas.
molecular_formulas <- c("C3F7O3S", "C4F9O3S", "C5F11O3S", "C6F9O3S", "C8F17O3S", "C9F19O3S", "C10F21O3S", "C7ClF14O4", "C10ClF20O4", "C11ClF22O4", "C11Cl2F21O4", "C12ClF24O4") ## ratio_delta_HBrClFI_C <- 2 # to aggregate polymeric classes mixed.HBrClFI.allowed <- FALSE # To detect only halogen saturated classes min_molecular_formula_class <- 2 max_number_formula_class <- 20 ## classes <- detect_formula_sets(molecular_formulas, ratio_delta_HBrClFI_C, mixed.HBrClFI.allowed, min_molecular_formula_class, max_number_formula_class, number_processing_threads = 1)molecular_formulas <- c("C3F7O3S", "C4F9O3S", "C5F11O3S", "C6F9O3S", "C8F17O3S", "C9F19O3S", "C10F21O3S", "C7ClF14O4", "C10ClF20O4", "C11ClF22O4", "C11Cl2F21O4", "C12ClF24O4") ## ratio_delta_HBrClFI_C <- 2 # to aggregate polymeric classes mixed.HBrClFI.allowed <- FALSE # To detect only halogen saturated classes min_molecular_formula_class <- 2 max_number_formula_class <- 20 ## classes <- detect_formula_sets(molecular_formulas, ratio_delta_HBrClFI_C, mixed.HBrClFI.allowed, min_molecular_formula_class, max_number_formula_class, number_processing_threads = 1)
This module sorts 84 non-labeled and 14 labeled elements in the periodic table for molecular formula deconvolution and isotopic profile calculation.
element_sorter(ElementList = "all", alphabeticalOrder = TRUE)element_sorter(ElementList = "all", alphabeticalOrder = TRUE)
ElementList |
A string vector of elements needed for isotopic profile calculation. The default value for this parameter is a vector string of entire elements. |
alphabeticalOrder |
'TRUE' should be used to sort the elements for elemental deconvolution (default value), 'FALSE' should be used to keep the input order. |
Elements |
A string vector of elements (alphabetically sorted or unsorted) |
massAbundanceList |
A list of isotopic mass and abundance of elements. |
Valence |
A vector of electron valences. |
EL_mass_abundance_val <- element_sorter()EL_mass_abundance_val <- element_sorter()
This function checks whether a molecular formula follows the extended SENIOR rule.
extendedSENIORrule(mol_vec, valence_vec, ionization_correction = 0)extendedSENIORrule(mol_vec, valence_vec, ionization_correction = 0)
mol_vec |
A vector of the deconvoluted molecular formula |
valence_vec |
A vector of the valences from the molecular formula. Valences may be acquired from the 'IUPAC_Isotopes' data. |
ionization_correction |
A number to compensate for the ionization losses/gains. For example, ‘-1' for [M+H/K/Na] ionization pathways and '+1’ for [M-H] ionization pathway. |
rule2 |
TURE for when the molecular formula passes the rule and FALSE for when the molecular formula fails to pass the rule. |
This function takes a formula and a vector of ionization pathways and returns the adduct formulas.
formula_adduct_calculator(molecular_formula, IonPathways)formula_adduct_calculator(molecular_formula, IonPathways)
molecular_formula |
molecular formula |
IonPathways |
A vector of ionization pathways. Pathways should be like [Coeff*M+ADD1-DED1+...] where "Coeff" should be an integer between 1-9 and ADD1 and DED1 may be ionization pathways. ex: 'IonPathways <- c("[M]+", "[M+H]+", "[2M-Cl]-", "[3M+CO2-H2O+Na-KO2+HCl-NH4]-")' |
A vector of adduct formulas
molecular_formula = "C15H10O7" IonPathways = c("[M+]","[M+H]","[M+H20+H]","[M+Na]") Formula_adducts <- formula_adduct_calculator(molecular_formula, IonPathways)molecular_formula = "C15H10O7" IonPathways = c("[M+]","[M+H]","[M+H20+H]","[M+Na]") Formula_adducts <- formula_adduct_calculator(molecular_formula, IonPathways)
This function convert a molecular formulas into a numerical vector
formula_vector_generator(molecular_formula, Elements, LElements = length(Elements), allowedRedundantElements = FALSE)formula_vector_generator(molecular_formula, Elements, LElements = length(Elements), allowedRedundantElements = FALSE)
molecular_formula |
molecular formula |
Elements |
a string vector of elements. This value must be driven from the 'element_sorter' function. |
LElements |
number of elements. To speed up loop calculations, consider calculating the number of elements outside of the loop. |
allowedRedundantElements |
'TRUE' should be used to deconvolute molecular formulas with redundant elements (e.g. CO2CH3O), and 'FALSE' should be used to skip such complex molecular formulas.(default value) |
a numerical vector for the molecular formula. This function returns a vector of -Inf values when the molecular formula has elements not listed in the 'Elements' string vector.
molecular_formula <- "[13]C2C12H2Br5Cl3O" Elements_molecular_formula <- c("[13]C", "C", "H", "O", "Br", "Cl") EL <- element_sorter(ElementList = Elements_molecular_formula, alphabeticalOrder = TRUE) Elements <- EL[["Elements"]] LElements <- length(Elements) ## mol_vec <- formula_vector_generator(molecular_formula, Elements, LElements, allowedRedundantElements = TRUE) ## regenerated_molecular_formula <- hill_molecular_formula_printer(Elements, mol_vec)molecular_formula <- "[13]C2C12H2Br5Cl3O" Elements_molecular_formula <- c("[13]C", "C", "H", "O", "Br", "Cl") EL <- element_sorter(ElementList = Elements_molecular_formula, alphabeticalOrder = TRUE) Elements <- EL[["Elements"]] LElements <- length(Elements) ## mol_vec <- formula_vector_generator(molecular_formula, Elements, LElements, allowedRedundantElements = TRUE) ## regenerated_molecular_formula <- hill_molecular_formula_printer(Elements, mol_vec)
This function produces molecular formulas from a list numerical vectors in the Hill notation system
hill_molecular_formula_printer(Elements, MolVecMat, number_processing_threads = 1)hill_molecular_formula_printer(Elements, MolVecMat, number_processing_threads = 1)
Elements |
A vector string of the used elements. |
MolVecMat |
A matrix of numerical vectors of molecular formulas in each row. |
number_processing_threads |
Number of processing threads for multi-threaded processing |
A vector of molecular formulas
Elements <- c("C", "H", "O", "N", "Br", "Cl") MoleFormVec1 <- c(2, 6, 1, 0, 0, 0) # C2H6O MoleFormVec2 <- c(8, 10, 2, 4, 0 ,0) # C8H10N4O2 MoleFormVec3 <- c(12, 2, 1, 0, 5, 3) # C12H2Br5Cl3O MolVecMat <- rbind(MoleFormVec1, MoleFormVec2, MoleFormVec3) H_MolF <- hill_molecular_formula_printer(Elements, MolVecMat)Elements <- c("C", "H", "O", "N", "Br", "Cl") MoleFormVec1 <- c(2, 6, 1, 0, 0, 0) # C2H6O MoleFormVec2 <- c(8, 10, 2, 4, 0 ,0) # C8H10N4O2 MoleFormVec3 <- c(12, 2, 1, 0, 5, 3) # C12H2Br5Cl3O MolVecMat <- rbind(MoleFormVec1, MoleFormVec2, MoleFormVec3) H_MolF <- hill_molecular_formula_printer(Elements, MolVecMat)
This function calculates the score values to rank candidate molecular formulas for a mass spectrometry-chromatography peak.
identificationScoreCalculator(scoreCoefficients, nIisotopologues, PCS, RCS, NEME, R13C_PL, R13C_IP)identificationScoreCalculator(scoreCoefficients, nIisotopologues, PCS, RCS, NEME, R13C_PL, R13C_IP)
scoreCoefficients |
A vector of seven numbers equal or greater than 0 |
nIisotopologues |
Number of isotopologues in the theoretical isotopic profiles. |
PCS |
PCS (per mille) |
RCS |
RCS (percentage) |
NEME |
NEME (mDa) |
R13C_PL |
R13C of the peak from IDSL.IPA peaklists |
R13C_IP |
R13C from theoretical isotopic profiles |
This function deconvolutes ionization pathways into a coefficient and a numerical vector to simplify prediction ionization pathways.
ionization_pathway_deconvoluter(IonPathways, Elements)ionization_pathway_deconvoluter(IonPathways, Elements)
IonPathways |
A vector of ionization pathways. Pathways should be like [Coeff*M+ADD1-DED1+...] where "Coeff" should be an integer between 1-9 and ADD1 and DED1 may be ionization pathways. ex: 'IonPathways <- c("[M]+", "[M+H]+", "[2M-Cl]-", "[3M+CO2-H2O+Na-KO2+HCl-NH4]-")' |
Elements |
A vector string of the used elements |
A list of adduct calculation values for each ionization pathway.
Elements <- element_sorter(alphabeticalOrder = TRUE)[["Elements"]] IonPathways <- c("[M]+", "[M+H]+", "[2M-Cl]-", "[3M+CO2-H2O+Na-KO2+HCl-NH4]-") Ion_DC <- ionization_pathway_deconvoluter(IonPathways, Elements)Elements <- element_sorter(alphabeticalOrder = TRUE)[["Elements"]] IonPathways <- c("[M]+", "[M+H]+", "[2M-Cl]-", "[3M+CO2-H2O+Na-KO2+HCl-NH4]-") Ion_DC <- ionization_pathway_deconvoluter(IonPathways, Elements)
This function was designed to calculate isotopic profile distributions for small molecules with masses <= 1200 Da. Nonetheless, this function may suit more complicated tasks with complex biological compounds. Details of the equations used in this function are available in the reference[1]. In this function, neighboring isotopologues are merged using the satellite clustering merging (SCM) method described in the reference[2].
isotopic_profile_calculator(MoleFormVec, massAbundanceList, peak_spacing, intensity_cutoff, UFA_IP_memeory_variables = c(1e30, 1e-12, 100))isotopic_profile_calculator(MoleFormVec, massAbundanceList, peak_spacing, intensity_cutoff, UFA_IP_memeory_variables = c(1e30, 1e-12, 100))
MoleFormVec |
A numerical vector of the molecular formula |
massAbundanceList |
A list of isotopic mass and abundance of elements obtained from the 'element_sorter' function |
peak_spacing |
A maximum space between two isotopologues in Da |
intensity_cutoff |
A minimum intensity threshold for isotopic profiles in percentage |
UFA_IP_memeory_variables |
A vector of three variables. Default values are c(1e30, 1e-12, 100) to manage memory usage. UFA_IP_memeory_variables[1] is used to control the overall size of isotopic combinations. UFA_IP_memeory_variables[2] indicates the minimum relative abundance (RA calculated by eq(1) in the reference [1]) of an isotopologue to include in the isotopic profile calculations. UFA_IP_memeory_variables[3] is the maximum elapsed time to calculate the isotopic profile on the 'setTimeLimit' function of base R. |
A matrix of isotopic profile. The first and second column represents the mass and intensity profiles, respectively.
[1] Fakouri Baygi, S., Crimmins, B.S., Hopke, P.K. Holsen, T.M. (2016). Comprehensive emerging chemical discovery: novel polyfluorinated compounds in Lake Michigan trout. Environmental Science and Technology, 50(17), 9460-9468, doi:10.1021/acs.est.6b01349.
[2] Fakouri Baygi, S., Fernando, S., Hopke, P.K., Holsen, T.M. and Crimmins, B.S. (2019). Automated Isotopic Profile Deconvolution for High Resolution Mass Spectrometric Data (APGC-QToF) from Biological Matrices. Analytical chemistry, 91(24), 15509-15517, doi:10.1021/acs.analchem.9b03335.
EL <- element_sorter(alphabeticalOrder = TRUE) Elements <- EL[["Elements"]] massAbundanceList <- EL[["massAbundanceList"]] peak_spacing <- 0.005 # mDa intensity_cutoff <- 1 # (in percentage) MoleFormVec <- formula_vector_generator("C8H10N4O2", Elements) IP <- isotopic_profile_calculator(MoleFormVec, massAbundanceList, peak_spacing, intensity_cutoff)EL <- element_sorter(alphabeticalOrder = TRUE) Elements <- EL[["Elements"]] massAbundanceList <- EL[["massAbundanceList"]] peak_spacing <- 0.005 # mDa intensity_cutoff <- 1 # (in percentage) MoleFormVec <- formula_vector_generator("C8H10N4O2", Elements) IP <- isotopic_profile_calculator(MoleFormVec, massAbundanceList, peak_spacing, intensity_cutoff)
This module annotates candidate molecular formulas in the peaklists from the IDSL.IPA pipeline using isotopic profiles.
molecular_formula_annotator(IPDB, spectraList, peaklist, selectedIPApeaks, massAccuracy, maxNEME, minPCS, minNDCS, minRCS, scoreCoefficients, RTtolerance = NA, correctedRTpeaklist = NULL, exportSpectraParameters = NULL, number_processing_threads = 1)molecular_formula_annotator(IPDB, spectraList, peaklist, selectedIPApeaks, massAccuracy, maxNEME, minPCS, minNDCS, minRCS, scoreCoefficients, RTtolerance = NA, correctedRTpeaklist = NULL, exportSpectraParameters = NULL, number_processing_threads = 1)
IPDB |
An isotopic profile database produced by the IDSL.UFA functions. |
spectraList |
a list of mass spectra in each chromatogram scan. |
peaklist |
Peaklist from the IDSL.IPA pipeline |
selectedIPApeaks |
selected IPA peaklist |
massAccuracy |
Mass accuracy in Da |
maxNEME |
Maximum value for Normalized Euclidean Mass Error (NEME) in mDa |
minPCS |
Minimum value for Profile Cosine Similarity (PCS) |
minNDCS |
Minimum value for Number of Detected Chromatogram Scans (NDCS) |
minRCS |
Minimum value for Ratio of Chromatogram Scans (RCS) in percentage |
scoreCoefficients |
A vector of five numbers representing coefficients of the identification score |
RTtolerance |
Retention time tolerance (min) |
correctedRTpeaklist |
corrected retention time peaklist |
exportSpectraParameters |
Parameters for export MS/MS match figures |
number_processing_threads |
Number of processing threads for multi-threaded processing |
A dataframe of candidate molecular formulas
molecular_formula_elements_filter
molecular_formula_elements_filter(molecularFormulaMatrix, Elements)molecular_formula_elements_filter(molecularFormulaMatrix, Elements)
molecularFormulaMatrix |
molecularFormulaMatrix |
Elements |
Elements |
a list of molecularFormulaMatrix and elementSorterList.
This function generates an efficient database for molecular formula matching against a database.
molecular_formula_library_generator(entire_molecular_formulas)molecular_formula_library_generator(entire_molecular_formulas)
entire_molecular_formulas |
A string vector of molecular formulas (redundancy is allowed) |
A vector of frequency of molecular formulas in the database.
entire_molecular_formulas <- c("C2H6O", "C2H6O", "C2H6O", "C2H6O", "CH4O", "CH4O", "CH4O", "NH4", "C6H12O6") db <- molecular_formula_library_generator(entire_molecular_formulas) freq <- db[c("C6H12O6", "CH4O")]entire_molecular_formulas <- c("C2H6O", "C2H6O", "C2H6O", "C2H6O", "CH4O", "CH4O", "CH4O", "NH4", "C6H12O6") db <- molecular_formula_library_generator(entire_molecular_formulas) freq <- db[c("C6H12O6", "CH4O")]
This function attempts to match candidate molecular formulas against a library of molecular formulas using a set of ionization pathways.
molecular_formula_library_search(MolecularFormulaAnnotationTable, MFlibrary, IonPathways, number_processing_threads = 1)molecular_formula_library_search(MolecularFormulaAnnotationTable, MFlibrary, IonPathways, number_processing_threads = 1)
MolecularFormulaAnnotationTable |
A molecular formula annotation table from the 'molecular_formula_annotator' module. |
MFlibrary |
A library of molecular formulas generated using the 'molecular_formula_library_generator' module. |
IonPathways |
A vector of ionization pathways. Pathways should be like [Coeff*M+ADD1-DED1+...] where "Coeff" should be an integer between 1-9 and ADD1 and DED1 may be ionization pathways. ex: 'IonPathways <- c("[M]+", "[M+H]+", "[2M-Cl]-", "[3M+CO2-H2O+Na-KO2+HCl-NH4]-")' |
number_processing_threads |
Number of processing threads for multi-threaded processing |
A function to calculate IPDBs from a vector of molecular formulas
molecularFormula2IPdb(molecularFormulaDatabase, retentionTime = NULL, peak_spacing = 0, intensity_cutoff_str = 1, IonPathways = "[M]+", number_processing_threads = 1, UFA_IP_memeory_variables = c(1e30, 1e-12, 100), allowedMustRunCalculation = FALSE, allowedVerbose = TRUE)molecularFormula2IPdb(molecularFormulaDatabase, retentionTime = NULL, peak_spacing = 0, intensity_cutoff_str = 1, IonPathways = "[M]+", number_processing_threads = 1, UFA_IP_memeory_variables = c(1e30, 1e-12, 100), allowedMustRunCalculation = FALSE, allowedVerbose = TRUE)
molecularFormulaDatabase |
A vector string of molecular formulas OR a list of elements and molecular formula matrix |
retentionTime |
retention time |
peak_spacing |
A maximum space between isotopologues in Da to merge neighboring isotopologues. |
intensity_cutoff_str |
A minimum intensity threshold for isotopic profiles in percentage. This parameter may be a string piece of R commands using c, b, br, cl, k, s, se, and si variables corresponding to the same elements. |
IonPathways |
A vector of ionization pathways. Pathways should be like [Coeff*M+ADD1-DED1+...] where "Coeff" should be an integer between 1-9 and ADD1 and DED1 may be ionization pathways. ex: 'IonPathways <- c("[M]+", "[M+H]+", "[2M-Cl]-", "[3M+CO2-H2O+Na-KO2+HCl-NH4]-")' |
number_processing_threads |
number of processing cores for multi-threaded computations. |
UFA_IP_memeory_variables |
A vector of three variables. Default values are c(1e30, 1e-12, 100) to manage memory usage. UFA_IP_memeory_variables[1] is used to control the overall size of isotopic combinations. UFA_IP_memeory_variables[2] indicates the minimum relative abundance (RA calculated by eq(1) in the reference [1]) of an isotopologue to include in the isotopic profile calculations. UFA_IP_memeory_variables[3] is the maximum elapsed time to calculate the isotopic profile on the 'setTimeLimit' function of the base R. |
allowedMustRunCalculation |
c(TRUE, FALSE). A 'TRUE' allowedMustRunCalculation applies a brute-force method to calculate complex isotopic profiles. When 'TRUE', this option may significantly reduce the speed for multithreaded processing. |
allowedVerbose |
c(TRUE, FALSE). A 'TRUE' allowedVerbose provides messages about the flow of the function. |
An IPDB list of isotopic profiles
[1] Fakouri Baygi, S., Crimmins, B.S., Hopke, P.K. Holsen, T.M. (2016). Comprehensive emerging chemical discovery: novel polyfluorinated compounds in Lake Michigan trout. Environmental Science and Technology, 50(17), 9460-9468, doi:10.1021/acs.est.6b01349.
library(IDSL.UFA, attach.required = TRUE) molecular_formula <- c("C13F8N8O2", "C20H22", "C8HF16ClSO3", "C12Cl10", "C123H193N35O37") peak_spacing <- 0.005 # in Da for QToF instruments # Use this piece of code for intensity cutoff to preserve significant isotopologues intensity_cutoff_str <- "if (s>0 & si>0) {min(c(c, 10, si*3, s*4))} else if (s>0 & si==0) {min(c(c, 10, s*4))} else if (s==0 & si>0) {min(c(c, 10, si*3))} else if (s==0 & si==0) {min(c(c, 10))}" UFA_IP_memeory_variables <- c(1e30, 1e-12, 100) IonPathways <- c("[M+H]+", "[M+Na]+", "[M-H2O+H]+") number_processing_threads <- 2 listIsoProDataBase <- molecularFormula2IPdb(molecular_formula, retentionTime = NULL, peak_spacing, intensity_cutoff_str, IonPathways, number_processing_threads, UFA_IP_memeory_variables, allowedMustRunCalculation = FALSE, allowedVerbose = TRUE) save(listIsoProDataBase, file = "listIsoProDataBase.Rdata")library(IDSL.UFA, attach.required = TRUE) molecular_formula <- c("C13F8N8O2", "C20H22", "C8HF16ClSO3", "C12Cl10", "C123H193N35O37") peak_spacing <- 0.005 # in Da for QToF instruments # Use this piece of code for intensity cutoff to preserve significant isotopologues intensity_cutoff_str <- "if (s>0 & si>0) {min(c(c, 10, si*3, s*4))} else if (s>0 & si==0) {min(c(c, 10, s*4))} else if (s==0 & si>0) {min(c(c, 10, si*3))} else if (s==0 & si==0) {min(c(c, 10))}" UFA_IP_memeory_variables <- c(1e30, 1e-12, 100) IonPathways <- c("[M+H]+", "[M+Na]+", "[M-H2O+H]+") number_processing_threads <- 2 listIsoProDataBase <- molecularFormula2IPdb(molecular_formula, retentionTime = NULL, peak_spacing, intensity_cutoff_str, IonPathways, number_processing_threads, UFA_IP_memeory_variables, allowedMustRunCalculation = FALSE, allowedVerbose = TRUE) save(listIsoProDataBase, file = "listIsoProDataBase.Rdata")
This function calculates monoisotopic mass of a molecular formula
monoisotopicMassCalculator(MoleFormVec, massAbundanceList, LElements = length(massAbundanceList))monoisotopicMassCalculator(MoleFormVec, massAbundanceList, LElements = length(massAbundanceList))
MoleFormVec |
A numerical vector molecular formula |
massAbundanceList |
A list of isotopic mass and abundance of elements obtained from the 'element_sorter' function |
LElements |
length of elements |
The monoisotopic mass
Elements <- c("C", "H", "O") MoleFormVec <- c(2, 6, 1) # C2H6O EL_mass_abundance <- element_sorter(ElementList = Elements, alphabeticalOrder = FALSE) massAbundanceList <- EL_mass_abundance[["massAbundanceList"]] MImass <- monoisotopicMassCalculator(MoleFormVec, massAbundanceList)Elements <- c("C", "H", "O") MoleFormVec <- c(2, 6, 1) # C2H6O EL_mass_abundance <- element_sorter(ElementList = Elements, alphabeticalOrder = FALSE) massAbundanceList <- EL_mass_abundance[["massAbundanceList"]] MImass <- monoisotopicMassCalculator(MoleFormVec, massAbundanceList)
This function evaluates the efficiency of the optimization process.
scoreCoefficientsEvaluation(PARAM_ScoreFunc)scoreCoefficientsEvaluation(PARAM_ScoreFunc)
PARAM_ScoreFunc |
PARAM_ScoreFunc is a variable derived from the 'UFA_coefficient_xlsxAnalyzer' function |
This function optimizes the coefficients of the score function.
scoreCoefficientsOptimization(PARAM_ScoreFunc)scoreCoefficientsOptimization(PARAM_ScoreFunc)
PARAM_ScoreFunc |
PARAM_ScoreFunc is a variable derived from the 'UFA_score_function_optimization_xlsxAnalyzer' function |
This function generates the input for the score optimization.
scoreCoefficientsReplicate(PARAM_ScoreFunc)scoreCoefficientsReplicate(PARAM_ScoreFunc)
PARAM_ScoreFunc |
PARAM_ScoreFunc is a variable derived from the 'UFA_coefficient_xlsxAnalyzer' function |
This function produces the isotopic profile database using the UFA enumerated chemical space (ECS) approach.
UFA_enumerated_chemical_space(PARAM_ECS)UFA_enumerated_chemical_space(PARAM_ECS)
PARAM_ECS |
A dataframe of the molecular formula constraints in the UFA spreadsheet |
This function evaluates the molecular formula generation constraints in the spreadsheet to create the isotopic profile database.
UFA_enumerated_chemical_space_xlsxAnalyzer(spreadsheet)UFA_enumerated_chemical_space_xlsxAnalyzer(spreadsheet)
spreadsheet |
UFA spreadsheet |
This function produces IPDB from a molecular formula source (a .csv/.txt/.xlsx file).
UFA_formula_source(PARAM_FormSource)UFA_formula_source(PARAM_FormSource)
PARAM_FormSource |
PARAM_FormSource is an internal variable of the IDSL.UFA package. |
an IPDB is saved in the destination address
This function evaluates the parameter spreadsheet for score coefficients optimization.
UFA_formula_source_xlsxAnalyzer(spreadsheet)UFA_formula_source_xlsxAnalyzer(spreadsheet)
spreadsheet |
The parameter spreadsheet in the .xlsx format. |
a processed parameter to feed the 'UFA_molecular_formulas_source' function.
To merge multiple IPDBs into one IPDB
UFA_IPdbMerger(path, vecIPDB)UFA_IPdbMerger(path, vecIPDB)
path |
path |
vecIPDB |
a vector of IPDBs |
IPDB
Locate indices of the pattern in the string
UFA_locate_regex(string, pattern, ignore.case = FALSE, perl = FALSE, fixed = FALSE, useBytes = FALSE)UFA_locate_regex(string, pattern, ignore.case = FALSE, perl = FALSE, fixed = FALSE, useBytes = FALSE)
string |
a string as character |
pattern |
a pattern to screen |
ignore.case |
ignore.case |
perl |
perl |
fixed |
fixed |
useBytes |
useBytes |
This function returns 'NULL' when no matches are detected for the pattern.
A 2-column matrix of location indices. The first and second columns represent start and end positions, respectively.
pattern <- "Cl" string <- "NaCl.5HCl" Location_Cl <- UFA_locate_regex(string, pattern)pattern <- "Cl" string <- "NaCl.5HCl" Location_Cl <- UFA_locate_regex(string, pattern)
This module is to extract molecular formulas from a database.
UFA_PubChem_formula_extraction(path)UFA_PubChem_formula_extraction(path)
path |
path to store information |
Molecular formulas in https://pubchem.ncbi.nlm.nih.gov/ are stored in the provided 'path' in the .txt format.
Kim S, Chen J, Cheng T, Gindulyte A, He J, He S, Li Q, Shoemaker BA, Thiessen PA, Yu B, Zaslavsky L, Zhang J, Bolton EE. PubChem in 2021: new data content and improved web interfaces. Nucleic Acids Res., 2021 Jan 8; 49(D1):D1388–D1395 , doi:10.1093/nar/gkaa971.
This function updates ranking orders of the individual MolecularFormulaAnnotationTable when score coefficients changed.
UFA_score_coefficients_corrector(input_annotated_molf_address, output_annotated_molf_address, scoreCoefficients, number_processing_threads = 1)UFA_score_coefficients_corrector(input_annotated_molf_address, output_annotated_molf_address, scoreCoefficients, number_processing_threads = 1)
input_annotated_molf_address |
Address to load the individual MolecularFormulaAnnotationTables. |
output_annotated_molf_address |
Address to save the individual MolecularFormulaAnnotationTables. |
scoreCoefficients |
A vector of five numbers representing coefficients of the identification score function. |
number_processing_threads |
Number of processing threads for multi-threaded computations. |
This function runs the score optimization workflow.
UFA_score_function_optimization(PARAM_ScoreFunc)UFA_score_function_optimization(PARAM_ScoreFunc)
PARAM_ScoreFunc |
PARAM_ScoreFunc from the 'UFA_score_function_optimization_xlsxAnalyzer' module |
This function evaluates the parameter spreadsheet for score coefficients optimization.
UFA_score_function_optimization_xlsxAnalyzer(spreadsheet)UFA_score_function_optimization_xlsxAnalyzer(spreadsheet)
spreadsheet |
The parameter spreadsheet in the .xlsx format. |
a processed parameter to feed the 'UFA_score_function_optimization' function.
This function executes the UFA workflow in order.
UFA_workflow(spreadsheet)UFA_workflow(spreadsheet)
spreadsheet |
UFA spreadsheet |
This function organizes the UFA file processing for better performance using the template spreadsheet.
This function processes the spreadsheet of the UFA parameters to ensure the parameter inputs are consistent with the requirements of the IDSL.UFA pipeline.
UFA_xlsxAnalyzer(spreadsheet)UFA_xlsxAnalyzer(spreadsheet)
spreadsheet |
UFA spreadsheet |
This function returns the UFA parameters to feed the UFA_workflow function.