simonboothroyd / constructure Goto Github PK
View Code? Open in Web Editor NEWConstruct molecules from diverse scaffolds and fragments.
License: MIT License
constructure's People
Contributors
Stargazers
Watchers
Forkers
lilyminiumconstructure's Issues
Non-sequential R-groups
Must R-groups be sequential? e.g. if I have a molecule that I number in my head by carbon,
it is easier for me to keep numbering by carbon than transfer that to 1, 2, 3, 4. However, Constructor.enumerate_combinations validates substituents starting from R-group 1. Could the validation function just check the given scaffold for defined R-groups instead? I can implement this if you think this is a valuable addition to constructure.
e.g.
# Create the scaffold.
scaffold = Scaffold(
smiles="C1(=C([C]([R4])=C2C(=[C]([R7])1)[N]([C]([R2])=[C]([R3])2)[H])O)O",
r_groups={
2: sub_groups,
3: sub_groups,
4: sub_groups,
7: sub_groups,
},
)
smiles = Constructor.enumerate_combinations(scaffold,
substituents={
2: substituents,
3: substituents,
4: substituents,
7: substituents,
})Cannot parse R-groups > 9
At present constructor cannot parse SMILES strings with R-groups > 9, due to
- not capturing more than single digits, and
- excluding 0
Code:
>>> from constructure.scaffolds import Scaffold
>>> from constructure.constructors import RDKitConstructor as Constructor
>>> # this smiles has 12 R-groups numbered 1 to 12
>>> smiles = "C1(C2C(C(C(C1([R1]))([R2]))([R3]))C(C3C(C2([R4]))C(C4C(C3([R5]))C(C(C(C4([R6]))([R7]))([R8]))([R9]))([R10]))([R11]))([R12])"
>>> r_groups = ["hydrogen"]
>>> scaffold = Scaffold(smiles=smiles,
... r_groups={i+1: r_groups for i in range(12)})
>>> Constructor.n_replaceable_groups(scaffold)
RDKit ERROR: [21:17:35] SMILES Parse Error: syntax error while parsing: C1(C2C(C(C(C1([1*]))([2*]))([3*]))C(C3C(C2([4*]))C(C4C(C3([5*]))C(C(C(C4([6*]))([7*]))([8*]))([9*]))([R10]))([1*]))([2*])Changing:
--- re.sub(r"\(\[R([1-9])+]\)", r"([\1*])", scaffold.smiles)
+++ re.sub(r"\(\[R([0-9]+)]\)", r"([\1*])", scaffold.smiles)may help.
Cannot include stereochemistry
Stereochemistry around the R-groups is not accounted for in the current regex. The string I'm using now is scaffold_smiles = re.sub(r"\(([\\/]*)\[R([0-9]+)]\)", r"(\1[\2*])", scaffold.smiles). Due to RDKit reaction SMARTS not preserving stereochemistry in the one test case I have looked at (rdkit/rdkit#4059), it may also require enumeration of the potential stereoisomers in attach_substituents (as suggested by you on Slack). I haven't seen any edge cases in OpenEye so far.
Add note re: Python versions
Just as a note for future documentation, this requires Python >=3.8 because of from typing import Literal.
Couple bugs in smiles_to_image_grid with RDKit backend
overview
- Draw.MolsToGridImage takes keyword arguments, not positional
- Passing in
legend=None(from the current keyword argumentslabels=None) borks the RDKit Draw.MolsToGridImage where it tries to subscript the list Imagehas no attributesave, so it dies when smiles_to_image_grid tries toimage.save(output_path)
code
# Import the scaffold object which stores the scaffold definition.
from constructure.scaffolds import Scaffold
from constructure.constructors import RDKitConstructor as Constructor
from rdkit import Chem
sub_groups = ["hydrogen", "alkyl", "aryl", "acyl",
"hetero", "halogen"]
# Create the scaffold.
scaffold = Scaffold(
smiles="C1(=C([C]([R1])=C2C(=[C]([R2])1)[N]([C]([R3])=[C]([R4])2)[H])O)O",
r_groups={
1: sub_groups,
2: sub_groups,
3: sub_groups,
4: sub_groups,
},
)
substituents = ["[R][H]", # -H
"[R]C", # -CH3
"[R]C=O",
"[R]N(=O)(=O)",
"[R]O",
"[R]-F",
"[R]-Cl",
"[R]-Br",
"[R]C(=O)(O)",
]
smiles = Constructor.enumerate_combinations(scaffold,
substituents={
1: substituents,
2: substituents,
3: substituents,
4: substituents,
})
from constructure.utilities.rdkit import smiles_to_image_grid
smiles_to_image_grid(smiles, "generated.png", cols=4)error traces
- Keyword arguments
>>> smiles_to_image_grid(smiles, "generated.png", cols=4)
---------------------------------------------------------------------------
TypeError Traceback (most recent call last)
<ipython-input-7-107a6a4db137> in <module>
1 from constructure.utilities.rdkit import smiles_to_image_grid
----> 2 smiles_to_image_grid(smiles, "generated.png", cols=4)
~/pydev/constructure/constructure/utilities/utilities.py in wrapper(*args, **kwargs)
60 raise MissingOptionalDependency(import_path, False)
61
---> 62 return function(*args, **kwargs)
63
64 return wrapper
~/pydev/constructure/constructure/utilities/rdkit.py in smiles_to_image_grid(smiles, output_path, labels, cols, cell_width, cell_height)
30 molecules = [Chem.MolFromSmiles(pattern) for pattern in smiles]
31
---> 32 image = Draw.MolsToGridImage(molecules, cols, (cell_width, cell_height), labels)
33 image.save(output_path)
34
TypeError: ShowMols() takes from 1 to 2 positional arguments but 4 were given- Legend=None
>>> smiles_to_image_grid(smiles, "generated.png", cols=4)
---------------------------------------------------------------------------
TypeError Traceback (most recent call last)
<ipython-input-11-107a6a4db137> in <module>
1 from constructure.utilities.rdkit import smiles_to_image_grid
----> 2 smiles_to_image_grid(smiles, "generated.png", cols=4)
~/pydev/constructure/constructure/utilities/utilities.py in wrapper(*args, **kwargs)
60 raise MissingOptionalDependency(import_path, False)
61
---> 62 return function(*args, **kwargs)
63
64 return wrapper
~/pydev/constructure/constructure/utilities/rdkit.py in smiles_to_image_grid(smiles, output_path, labels, cols, cell_width, cell_height)
30 molecules = [Chem.MolFromSmiles(pattern) for pattern in smiles]
31
---> 32 image = Draw.MolsToGridImage(molecules, molsPerRow=cols, subImgSize=(cell_width, cell_height), legends=labels)
33 image.save(output_path)
34
~/anaconda3/envs/constructure/lib/python3.8/site-packages/rdkit/Chem/Draw/IPythonConsole.py in ShowMols(mols, maxMols, **kwargs)
194 for prop in ('legends', 'highlightAtoms', 'highlightBonds'):
195 if prop in kwargs:
--> 196 kwargs[prop] = kwargs[prop][:maxMols]
197 res = fn(mols, drawOptions=drawOptions, **kwargs)
198 if kwargs['useSVG']:
TypeError: 'NoneType' object is not subscriptable- No attribute save
smiles_to_image_grid(smiles, "generated.png", cols=4, labels=[])
---------------------------------------------------------------------------
AttributeError Traceback (most recent call last)
<ipython-input-5-131e1400cb77> in <module>
1 from constructure.utilities.rdkit import smiles_to_image_grid
----> 2 smiles_to_image_grid(smiles, "generated.png", cols=4, labels=[])
~/pydev/constructure/constructure/utilities/utilities.py in wrapper(*args, **kwargs)
60 raise MissingOptionalDependency(import_path, False)
61
---> 62 return function(*args, **kwargs)
63
64 return wrapper
~/pydev/constructure/constructure/utilities/rdkit.py in smiles_to_image_grid(smiles, output_path, labels, cols, cell_width, cell_height)
31
32 image = Draw.MolsToGridImage(molecules, molsPerRow=cols, subImgSize=(cell_width, cell_height), legends=labels)
---> 33 image.save(output_path)
34
35
AttributeError: 'Image' object has no attribute 'save'
versions
- Python: 3.8, 3.9
- RDKit: 2021.03.1
- constructure: current main
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.