Move SMARTS strings into a dataset

[Aariq]

By moving the SMARTS strings in get_fx_groups() (

volcalc/R/get_fx_groups.R

Lines 100 to 141 in 80bed5b

	# *_pattern are SMARTS strings: https://www.daylight.com/dayhtml_tutorials/languages/smarts/smarts_examples.html
	carbon_dbl_bonds_pattern <- "C=C" #non-aromatic carbon double bonds
	CCCO_pattern <- "C(C=C[AR1])(=O)[AR1]" #C=C-C=O in a non-aromatic ring
	# ether_alkyl_pattern <- "[OD2]([C!R1])[C!R1]" #currently unused--ether_alkly calculated as total - other ethers
	ether_alicyclic_pattern <- "[OD2]([C!R0])[C!R0]"
	ether_aromatic_pattern <- "O(c)[C,c]" #only one of the carbons has to be aromatic
	nitro_pattern <- "[$([NX3](=O)=O),$([NX3+](=O)[O-])][!#8]"
	hydroxyl_aromatic_pattern <- "[OX2H]c"
	nitrate_pattern <- "[$([NX3](=[OX1])(=[OX1])O),$([NX3+]([OX1-])(=[OX1])O)]"
	#TODO need patterns for amines that don't pick up amides
	amine_primary_pattern <- "[NX3;H2;!$(NC=[!#6]);!$(NC#[!#6])][#6X4]"
	amine_secondary_pattern <- "[NX3H1!$(NC=[!#6])!$(NC#[!#6])]([#6X4])[#6X4]"
	amine_tertiary_pattern <- "[NX3H0!$(NC=[!#6])!$(NC#[!#6])]([#6X4])([#6X4])[#6X4]"
	amine_aromatic_pattern <- "[NX3;!$(NO)]c"
	amide_primary_pattern <- "[CX3;$([R0][#6]),$([H1R0])](=[OX1])[#7X3H2]"
	amide_secondary_pattern <- "[CX3;$([R0][#6]),$([H1R0])](=[OX1])[#7X3H1][#6;!$(C=[O,N,S])]"
	amide_tertiary_pattern <-
	"[CX3;$([R0][#6]),$([H1R0])](=[OX1])[#7X3H0]([#6;!$(C=[O,N,S])])[#6;!$(C=[O,N,S])]"
	# amide_total_pattern <- "[CX3;$([R0][#6]),$([H1R0])](=[OX1])[#7X3;$([H2]),$([H1][#6;!$(C=[O,N,S])]),$([#7]([#6;!$(C=[O,N,S])])[#6;!$(C=[O,N,S])])]"
	carbonylperoxynitrate_pattern <- "*C(=O)OO[N+1](=O)[O-1]"
	peroxide_pattern <- "[OX2D2][OX2D2]" #this captures carbonylperoxynitrates too
	hydroperoxide_pattern <- "[OX2][OX2H,OX1-]" #this captures peroxyacids too
	carbonylperoxyacid_pattern <- "[CX3;$([R0][#6]),$([H1R0])](=[OX1])[OX2][$([OX2H]),$([OX1-])]"
	nitroester_pattern <- "C(=O)(OC)C~[NX3](-,=[OX1])-,=[OX1]"
	# This captures OH groups on a ring that also has a nitro group (para, ortho, or meta). Need to correct aromatic hydroxyl count later.
	nitrophenol_pattern <-
	"[OX2H][$(c1ccccc1[$([NX3](=O)=O),$([NX3+](=O)[O-])]),$(c1cccc(c1)[$([NX3](=O)=O),$([NX3+](=O)[O-])]),$(c1ccc(cc1)[$([NX3](=O)=O),$([NX3+](=O)[O-])])]"
	phosphoric_acid_pattern <-
	"[$(P(=[OX1])([$([OX2H]),$([OX1-]),$([OX2]P)])([$([OX2H]),$([OX1-]),$([OX2]P)])[$([OX2H]),$([OX1-]),$([OX2]P)]),$([P+]([OX1-])([$([OX2H]),$([OX1-]),$([OX2]P)])([$([OX2H]),$([OX1-]),$([OX2]P)])[$([OX2H]),$([OX1-]),$([OX2]P)])]"
	phosphoric_ester_pattern <-
	"[$(P(=[OX1])([OX2][#6])([$([OX2H]),$([OX1-]),$([OX2][#6])])[$([OX2H]),$([OX1-]),$([OX2][#6]),$([OX2]P)]),$([P+]([OX1-])([OX2][#6])([$([OX2H]),$([OX1-]),$([OX2][#6])])[$([OX2H]),$([OX1-]),$([OX2][#6]),$([OX2]P)])]"
	sulfate_pattern <-
	"[$([#16X4](=[OX1])(=[OX1])([OX2H,OX1H0-])[OX2][#6]),$([#16X4+2]([OX1-])([OX1-])([OX2H,OX1H0-])[OX2][#6])]"
	#sulfonate groups; sulfonate ions, and conjugate acid, sulfonic acids
	sulfonate_pattern <-
	"[#16X4](=[OX1])(=[OX1])([#6])[*$([O-1]),*$([OH1]),*$([OX2H0])]"
	thiol_pattern <- "[#16X2H]"
	carbothioester_pattern <- "S([#6])[CX3](=O)[#6]"

) into a dataframe included with the package, it would simplify the code, make the functional group definitions more transparent to users, and potentially make get_fx_groups() easier to expand to new methods besides SIMPOL.1.

Possible column names: method (?), functional_group, smarts, function. function would be useful for the groups that are captured by a ChemmineOB function other than a SMARTS search. For all the SMARTS strings, function would be "ChemmineR::smartsSearchOB"

[Aariq]

I feel like this is a superior option to #90 because it is less work and probably more useful.