Skip to content

Reaction fp #62

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
kochern wants to merge 11 commits into
base: master
Choose a base branch
from
Open

Reaction fp #62

Changes from 1 commit
Commits
Show all changes
11 commits
Select commit Hold shift + click to select a range
4305a35
added automatic number of bits to emulate count fps for binary format
Nov 18, 2025
5e5844f
added morgan_count_fingerprint
Nov 18, 2025
2827eb4
added reaction difference fingerprint
Nov 18, 2025
483dd00
Corrected how the binary fps are calculated. Added concatenation of r…
Nov 24, 2025
b767209
implementation of count fps via hash shift
Dec 2, 2025
55effbb
A finalized version of fingerprints. fragment counts are implemented …
Dec 11, 2025
70f7846
Merge branch 'master' into fork/kochern/reaction_fp
stsouko Jan 30, 2026
a32e84a
backward compatibility added.
stsouko Jan 30, 2026
9ca5f57
margan counter reimplemented
stsouko Mar 7, 2026
7647208
test update
stsouko Mar 7, 2026
6d4da30
diff fp refactored
stsouko Mar 7, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Next Next commit
added automatic number of bits to emulate count fps for binary format
  • Loading branch information
Kostiantyn Chernichenko
Kostiantyn Chernichenko committed Nov 18, 2025
commit 4305a35ad34ed001d4227cb75b118e098372b6a0
45 changes: 35 additions & 10 deletions chython/algorithms/fingerprints/morgan.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -40,7 +40,7 @@ def morgan_fingerprint(self, min_radius: int = 1, max_radius: int = 4,
:param min_radius: minimal radius of EC
:param max_radius: maximum radius of EC
:param length: bit string's length. Should be power of 2
:param number_active_bits: number of active bits for each hashed tuple
:param number_active_bits: number of active bits for each hashed tuple (int or 'auto'). For 'auto' option, number of bits is count of each hash + 1.

:return: array(n_features)
"""
Expand All @@ -57,20 +57,30 @@ def morgan_bit_set(self, min_radius: int = 1, max_radius: int = 4,
:param min_radius: minimal radius of EC
:param max_radius: maximum radius of EC
:param length: bit string's length. Should be power of 2
:param number_active_bits: number of active bits for each hashed tuple
:param number_active_bits: number of active bits for each hashed tuple (int or 'auto'). For 'auto' option, number of bits is count of each hash + 1.
"""
mask = length - 1
log = int(log2(length))

active_bits = set()
for tpl in self.morgan_hash_set(min_radius, max_radius):
active_bits.add(tpl & mask)
if number_active_bits == 2:
active_bits.add(tpl >> log & mask)
elif number_active_bits > 2:
for _ in range(1, number_active_bits):
tpl >>= log
active_bits.add(tpl & mask)

if number_active_bits == 'auto':
for hsh, cnt in self.morgan_hash_counts(min_radius, max_radius):
active_bits.add(hsh & mask)
active_bits.add(hsh >> log & mask)
for _ in range(cnt-1):
hsh >>= log
active_bits.add(hsh & mask)

else:
for tpl in self.morgan_hash_set(min_radius, max_radius):
active_bits.add(tpl & mask)
if number_active_bits == 2:
active_bits.add(tpl >> log & mask)
elif number_active_bits > 2:
for _ in range(1, number_active_bits):
tpl >>= log
active_bits.add(tpl & mask)
return active_bits

def morgan_hash_set(self: 'MoleculeContainer', min_radius: int = 1, max_radius: int = 4) -> Set[int]:
Expand Down Expand Up @@ -129,6 +139,21 @@ def _morgan_hash_dict(self: 'MoleculeContainer', min_radius: int = 1, max_radius
out.append(identifiers)
return out[-(max_radius - min_radius + 1):] # slice [min, max] radii range

def morgan_hash_counts(self, min_radius: int = 1, max_radius: int = 4) -> List[tuple]:
"""
Count occurrences of each hash from _morgan_hash_dict.

:param min_radius: minimal radius of EC
:param max_radius: maximum radius of EC
:return: list of (hash, count) tuples
"""
counts = defaultdict(int)
for hash_dict in self._morgan_hash_dict(min_radius, max_radius):
for h in hash_dict.values():
counts[h] += 1
return list(counts.items())


@property
def _atom_identifiers(self) -> Dict[int, int]:
raise NotImplementedError
Expand Down