conformer-generation

Name: conformer-generation
Author: GPTomics

1.1kstars185forks

How to install conformer-generation?

gptomics/bioskills/conformer-generation

$npx -y skills add GPTomics/bioSkills --skill conformer-generation

Installs into the current project.

Summary

Generates 3D conformer ensembles using RDKit ETKDGv3 with knowledge-enhanced distance geometry, MMFF94/UFF force-field optimization, CREST + GFN2-xTB semi-empirical refinement, and macrocycle-aware torsion preferences. Provides explicit decision rules for single vs ensemble confo

Files · 1

View on GitHub

SKILL.md

1	## Version Compatibility
2
3	Reference examples tested with: RDKit 2024.09+, xtb 6.7+, CREST 3.0+, OpenMM 8.1+ for follow-up MD.
4
5	Before using code patterns, verify installed versions match. If versions differ:
6	- Python: `pip show <package>` then `help(module.function)` to check signatures
7	- CLI: `xtb --version`; `crest --version`
8
9	If code throws ImportError, AttributeError, or TypeError, introspect the installed
10	package and adapt the example to match the actual API rather than retrying.
11
12	# Conformer Generation
13
14	Generate 3D conformer ensembles for molecules from 2D structures. The choice of method depends on molecule size, flexibility, and downstream use: ETKDG (Riniker & Landrum 2015) and its ETKDGv3 macrocycle update (Wang et al. 2020) are modern defaults for drug-like molecules, MMFF94/UFF provide fast energy minimization, and CREST + GFN2-xTB provide higher-cost semi-empirical sampling. A single conformer may be insufficient when the downstream result is conformation-sensitive; determine ensemble size by convergence of the downstream descriptor, alignment, or docking result.
15
16	For docking pose validation, see `chemoinformatics/pose-validation`. For free-energy methods (which require ensemble sampling), see `chemoinformatics/free-energy-calculations`.
17
18	## Conformer Method Taxonomy
19
20	\| Method \| Cost / mol \| Quality \| Use case \| Fails when \|
21	\|--------\|-----------\|---------\|----------\|------------\|
22	\| ETKDGv3 + MMFF94 \| Benchmark on actual molecules/hardware \| Useful for many drug-like organics \| Initial docking/descriptors \| Difficult macrocycles, peptides, unsupported chemistry \|
23	\| ETKDGv3 + UFF \| Fast \| Different parameter coverage from MMFF94 \| Fallback only after checking UFF parameters \| Unsupported atom types; coordination chemistry \|
24	\| Omega (OpenEye) \| Benchmark licensed workflow \| Commercial conformer generator \| Commercial pipelines \| License cost and configured limits \|
25	\| Confab (Open Babel) \| Benchmark on intended chemistry \| Systematic torsion search \| Alternative enumeration \| Combinatorial growth and force-field dependence \|
26	\| RDKit ETKDGv3 + macrocycle preferences \| Molecule-dependent \| Macrocycle-aware embedding \| Macrocyclic starting ensembles \| Coverage remains molecule-dependent \|
27	\| CREST + GFN2-xTB \| Molecule/settings-dependent \| Semiempirical conformational sampling \| Difficult flexible molecules \| Computational cost; special chemistry \|
28	\| CREST + GFN-FF \| Lower cost than GFN2-xTB \| Force-field-level sampling \| Exploratory sampling \| Validate coverage and ordering for the chemistry \|
29	\| GeoMol (Ganea 2021) \| Hardware/model-dependent \| Learned conformer generation \| Large-library research workflow \| Training distribution and released-model coverage \|
30	\| TorsionNet (Gogineni 2020) \| Hardware/model-dependent \| Learned torsional search \| Research workflow \| Training distribution and implementation availability \|
31	\| MD sampling (OpenMM) \| System/protocol-dependent \| Dynamic sampling \| Free energy, induced fit \| Computational cost and convergence \|
32
33	Decision: Start drug-like organic molecules with ETKDGv3 and a parameter-checked MMFF94/MMFF94s optimization. Escalate difficult macrocycles, peptides, or highly flexible molecules to a validated CREST workflow when downstream convergence is inadequate. Benchmark ML generators on the intended chemistry before using them at scale.
34
35	## Decision Tree by Scenario
36
37	\| Scenario \| Starting method \| Sampling and filtering decision \|
38	\|----------\|-----------------\|---------------------------------\|
39	\| Single initial 3D structure \| ETKDGv3 + checked force field \| Confirm embedding and minimization; downstream relaxation may still be required \|
40	\| Multi-conformer docking \| ETKDGv3 ensemble \| Increase sampling until pose recovery or enrichment is stable \|
41	\| 3D descriptors / pharmacophores \| ETKDGv3 ensemble \| Converge the reported statistic; justify energy/RMSD filters \|
42	\| Macrocycle / peptide \| Macrocycle-aware ETKDG, then CREST if needed \| Compare coverage against known conformers or downstream convergence \|
43	\| FEP input \| Bound-pose-informed preparation and MD \| Do not select solely by isolated-molecule conformer energy \|
44	\| Shape search \| Query- and library-specific ensemble \| Converge retrieval performance on a reference set \|
45
46	## ETKDGv3 (Modern Default)
47
48	ETKDGv3 (Wang et al. 2020), building on the original ETKDG method (Riniker & Landrum 2015), incorporates experimental torsio

People who liked this also viewed

6 picks

Type

conformer-generation

1.1kstars185forks

View on GitHub

How to install conformer-generation?

gptomics/bioskills/conformer-generation

$npx -y skills add GPTomics/bioSkills --skill conformer-generation

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	## Version Compatibility
2
3	Reference examples tested with: RDKit 2024.09+, xtb 6.7+, CREST 3.0+, OpenMM 8.1+ for follow-up MD.
4
5	Before using code patterns, verify installed versions match. If versions differ:
6	- Python: `pip show <package>` then `help(module.function)` to check signatures
7	- CLI: `xtb --version`; `crest --version`
8
9	If code throws ImportError, AttributeError, or TypeError, introspect the installed
10	package and adapt the example to match the actual API rather than retrying.
11
12	# Conformer Generation
13
14	Generate 3D conformer ensembles for molecules from 2D structures. The choice of method depends on molecule size, flexibility, and downstream use: ETKDG (Riniker & Landrum 2015) and its ETKDGv3 macrocycle update (Wang et al. 2020) are modern defaults for drug-like molecules, MMFF94/UFF provide fast energy minimization, and CREST + GFN2-xTB provide higher-cost semi-empirical sampling. A single conformer may be insufficient when the downstream result is conformation-sensitive; determine ensemble size by convergence of the downstream descriptor, alignment, or docking result.
15
16	For docking pose validation, see `chemoinformatics/pose-validation`. For free-energy methods (which require ensemble sampling), see `chemoinformatics/free-energy-calculations`.
17
18	## Conformer Method Taxonomy
19
20	\| Method \| Cost / mol \| Quality \| Use case \| Fails when \|
21	\|--------\|-----------\|---------\|----------\|------------\|
22	\| ETKDGv3 + MMFF94 \| Benchmark on actual molecules/hardware \| Useful for many drug-like organics \| Initial docking/descriptors \| Difficult macrocycles, peptides, unsupported chemistry \|
23	\| ETKDGv3 + UFF \| Fast \| Different parameter coverage from MMFF94 \| Fallback only after checking UFF parameters \| Unsupported atom types; coordination chemistry \|
24	\| Omega (OpenEye) \| Benchmark licensed workflow \| Commercial conformer generator \| Commercial pipelines \| License cost and configured limits \|
25	\| Confab (Open Babel) \| Benchmark on intended chemistry \| Systematic torsion search \| Alternative enumeration \| Combinatorial growth and force-field dependence \|
26	\| RDKit ETKDGv3 + macrocycle preferences \| Molecule-dependent \| Macrocycle-aware embedding \| Macrocyclic starting ensembles \| Coverage remains molecule-dependent \|
27	\| CREST + GFN2-xTB \| Molecule/settings-dependent \| Semiempirical conformational sampling \| Difficult flexible molecules \| Computational cost; special chemistry \|
28	\| CREST + GFN-FF \| Lower cost than GFN2-xTB \| Force-field-level sampling \| Exploratory sampling \| Validate coverage and ordering for the chemistry \|
29	\| GeoMol (Ganea 2021) \| Hardware/model-dependent \| Learned conformer generation \| Large-library research workflow \| Training distribution and released-model coverage \|
30	\| TorsionNet (Gogineni 2020) \| Hardware/model-dependent \| Learned torsional search \| Research workflow \| Training distribution and implementation availability \|
31	\| MD sampling (OpenMM) \| System/protocol-dependent \| Dynamic sampling \| Free energy, induced fit \| Computational cost and convergence \|
32
33	Decision: Start drug-like organic molecules with ETKDGv3 and a parameter-checked MMFF94/MMFF94s optimization. Escalate difficult macrocycles, peptides, or highly flexible molecules to a validated CREST workflow when downstream convergence is inadequate. Benchmark ML generators on the intended chemistry before using them at scale.
34
35	## Decision Tree by Scenario
36
37	\| Scenario \| Starting method \| Sampling and filtering decision \|
38	\|----------\|-----------------\|---------------------------------\|
39	\| Single initial 3D structure \| ETKDGv3 + checked force field \| Confirm embedding and minimization; downstream relaxation may still be required \|
40	\| Multi-conformer docking \| ETKDGv3 ensemble \| Increase sampling until pose recovery or enrichment is stable \|
41	\| 3D descriptors / pharmacophores \| ETKDGv3 ensemble \| Converge the reported statistic; justify energy/RMSD filters \|
42	\| Macrocycle / peptide \| Macrocycle-aware ETKDG, then CREST if needed \| Compare coverage against known conformers or downstream convergence \|
43	\| FEP input \| Bound-pose-informed preparation and MD \| Do not select solely by isolated-molecule conformer energy \|
44	\| Shape search \| Query- and library-specific ensemble \| Converge retrieval performance on a reference set \|
45
46	## ETKDGv3 (Modern Default)
47
48	ETKDGv3 (Wang et al. 2020), building on the original ETKDG method (Riniker & Landrum 2015), incorporates experimental torsio

People who liked this also viewed

6 picks

Type