canvasFPGen

This program generates fingerprints for a set of molecules. A variety of schemes for atom and bond typing are available, as listed in Table 1. There are also a number of scaling schemes, for the fingerprint values, listed in Table 2. See canvasFPGen Command Help for syntax and options.

Table 1. Atom typing schemes for the -atomtype option of the canvasFPGen command
Scheme	Description
1	All atoms equivalent; all bonds equivalent.
2	Atoms distinguished as hydrogen bond acceptors or donors; all bonds equivalent.
3	Atoms distinguished by hybridization state; all bonds equivalent.
4	Atoms distinguished by functional type: {H}, {C}, {F,Cl}, {Br,I}, {N,O}, {S}, {other}; bonds by hybridization.
5	Mol2 atom types; all bonds equivalent.
6	Atoms distinguished by whether they are terminal, halogens, hydrogen bond acceptors or donors; bonds distinguished by bond order.
7	Atomic number and bond order.
8	Atoms distinguished by ring size, aromaticity, hydrogen bond acceptor or donor, ionization potential, whether terminal or halogen; bonds distinguished by bond order.
9	Carhart atom types (atom-pairs approach); all bonds equivalent.
10	Daylight invariant atom types; bonds distinguished by bond order.
11	Atoms distinguished by atomic number and bond order, aromaticity.
12	Atoms distinguished by atomic number and bond order, aromaticity, if aliphatic whether cyclic or acyclic.
C	Custom. Must be followed by location of a type definitions file. See the definition files (`.typ`) in `$SCHRODINGER/mmshare-vversion/data/canvas` for examples. Cannot be used with fingerprint types `maccs` and `custom`.
E	Estate atom types. Cannot be used with fingerprint types `maccs` and `custom`.

Table 2. Fingerprint scaling options
Option	Description
0	No scaling (default)
1	Scale counts by feature size to unity
2	Scale counts by feature size to feature size
3	Scale counts by feature size to molecule size
4	Scale squares of counts by feature size to unity
5	Scale squares of counts by feature size to feature size
6	Scale squares of counts by feature size to molecule size
7	Scale square root of counts by feature size to unity
8	Scale square root of counts by feature size to feature size
9	Scale square root of counts by feature size to molecule size
10	Use raw feature counts
11	Use square of raw feature counts
12	Use square root of raw feature counts
13	Use constant value of one