ume

# Step 1: Assign formulas (checks the peaklist format and calculates neutral masses and mass accuracy)
  # calc_neutral_mass() and calc_ma_abs()
  mfd <- assign_formulas(pl = ume::peaklist_demo, formula_library = ume::lib_demo,
                        pol = "neg", ma_dev = 0.5, verbose = TRUE)
                        
# Step 2: Verify the existence of the major isotope signals and their magnitudes
  mfd <- eval_isotopes(mfd = mfd, remove_isotopes = TRUE, verbose = TRUE)

# Step 3: Calculate evaluation parameters
  mfd <- calc_eval_params(mfd = mfd, verbose = TRUE)

# Step 4: Add known classification for formulas
  # to do: the categories should be listed in one column containing the category assignment
  mfd <- add_known_mf(mfd = mfd)

# Step 5: Remove all formulas that occur in one or more blank analyses
  # The demo peaklist contains one blank spectrum named "Blank" (file_id = 1)
  # This removes all molecular formulas recorded in the blank from the entire dataset
  mfd <- remove_blanks(mfd = mfd, blank_file_ids = 1, blank_prevalence = 0)
  
# Step 6: Filter formula table according to evaluation parameters (generated in step 3)
  mfd_filt <- filter_mf_data(mfd = mfd, 
                             select_file_ids = 2:5,
                             dbe_o_max = 10,
                             oc_min = 0.2,
                             oc_max = 1.2,
                             verbose = TRUE)
    
# Step 7: Normalize intensities
  mfd_filt <- calc_norm_int(mfd = mfd_filt, normalization = "bp", verbose = TRUE)
  
# Step 8: Filter by (relative) peak magnitude (in this case: >= 5 percent base peak intensity)
  mfd_filt <- filter_int(mfd = mfd_filt, norm_int_min = 0.5, verbose = TRUE)

# Step 9: Normalize intensities
  mfd_filt <- calc_norm_int(mfd = mfd_filt, normalization = "bp", verbose = TRUE)
  
# Step 10: Order the columns of the results table  
  mfd_filt <- order_columns(mfd = mfd_filt)  

# Alternative using pipe operator:
  mf_data_demo |>
    eval_isotopes(remove_isotopes = T) |>
    calc_eval_params() |>
    add_known_mf() |>
    order_columns()

# Calculate double bond equivalent (DBE) for a molecular formula
# Uses isotope masses and element valences defined in ume::masses
# Calculation based on a molecular formula character vector
calc_dbe("C2H4")

# Based on a UME standard table:
data("mf_data_demo")
mf_data_demo[, dbe_new:= calc_dbe(mf_data_demo)]

# Nominal mass
calc_nm(mfd = c("C2[13C]H4", "C2H4", "C2H5OH", "C2H5OH"))

# Exact mass
calc_exact_mass(mfd = "C2[13C]H4")

# Neutral mass for (de-) protonated ions
calc_neutral_mass(123.1241, pol = "neg")

# Calculate mass accuracy
calc_ma(m = 228.0269, m_cal = 228.0270026)
calc_ma(m = 228.0269, m_cal = calc_exact_mass("C9H8O7"))

# Extract the molecular formula from an InChI code:
inchi_to_mf("InChI=1S/C2H6O/c1-2-3/h3H,2H2,1H3")

# Formula to table
convert_molecular_formula_to_data_table(mf = c("C2[13C]H4", "C2H5OH", "C2H6O"))

# In some specific cases the lightest isotope may not be the most abundant isotope.
# For these cases the function allows to choose between two interpretations of the 
# element symbol:
convert_molecular_formula_to_data_table("FeC10H10", isotope_default = "most_abundant")
convert_molecular_formula_to_data_table("FeC10H10", isotope_default = "lightest")

# Table to formula
dt <- convert_molecular_formula_to_data_table(mf = c("C2[13C]H4", "C2H5OH", "C2H6O"))

convert_data_table_to_molecular_formulas(mfd = dt, isotope_formulas = T)