feed · watching · stats · help with your tokens
asad/SMSD
A cheminformatics library for substructure, MCS, and fingerprint similarity search in C++, Java, and Python with optional GPU acceleration.
It solves the problem of finding substructures and maximum common substructures in molecular graphs for cheminformatics applications, providing consistent results across multiple programming languages. It offers a unified search model across C++, Java, and Python, supporting exact substructure, MCS, and fingerprint similarity with optional GPU acceleration.
Similar
- jyotipmahes/ML-Interview-Prep
A repository containing Machine Learning Engineer interview preparation materials and from-scratch algorithm implementations.
- postmanlabs/postman-app-support
The public support repository where Postman users file feature requests, integration ideas and bug reports.
- D4Vinci/Scrapling
Scrapling is a Python web scraping framework that combines fast HTTP requests, stealth browsers, parsing and concurrent crawling in one library.
- NVIDIA/auto-ontology
An open context layer that turns database metadata into an ontology, so AI agents answer questions about enterprise data with fewer wrong answers.
- storytold/craft-anything
Craft Anything is an open-source, pure Rust engine that studies software to understand its mechanics and uses AI agents to build new versions.
- soml-lang/soml
SOML is a configuration format for humans that keeps JSON's data model and strictness while dropping braces, quotes and commas.
- mamba-org/mamba
The Fast Cross-Platform Package Manager
- KKKKhazix/agent-autoloop
Agent Autoloop lets a command-line agent pick up work in the background and follow it through, asking for a decision only when something is uncertain.
sources
Claims below include README evidence. A matching quote or worker agreement does not establish that a claim is correct. Category majority among 2 workers. Version 1.
A cheminformatics library for substructure, MCS, and fingerprint similarity search in C++, Java, and Python with optional GPU acceleration.
SMSD Pro provides substructure search and maximum common substructure (MCS) search for C++, Java and Python, with the same search model in each language.
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
It solves the problem of finding substructures and maximum common substructures in molecular graphs for cheminformatics applications, providing consistent results across multiple programming languages.
SMSD Pro provides substructure search and maximum common substructure (MCS) search for C++, Java and Python, with the same search model in each language.
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
It offers a unified search model across C++, Java, and Python, supporting exact substructure, MCS, and fingerprint similarity with optional GPU acceleration.
SMSD Pro provides substructure search and maximum common substructure (MCS) search for C++, Java and Python, with the same search model in each language.
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Performs substructure searches using molecules or SMARTS queries, supporting both non-induced and induced embeddings.
Substructure search with molecule or SMARTS queries, as non-induced (default) or induced embeddings.
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Computes the maximum common substructure (MCS) as either the maximum common induced subgraph (MCIS) or the maximum common edge subgraph (MCES).
MCS search for the maximum common induced subgraph (MCIS) or the maximum common edge subgraph (MCES), connected or as disconnected fragments.
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Supports various chemistry options including bond order, aromaticity, formal charge, isotopes, ring membership, stereochemistry, and tautomers.
Chemistry options for bond order, aromaticity, formal charge, isotopes, ring membership, stereo and tautomers.
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Allows prepared queries to be efficiently searched against large collections of targets using multiple threads.
Prepared queries that can be searched against many targets, in turn or on several threads.
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Provides multiple fingerprint types for screening, including circular (ECFP/FCFP), path, topological-torsion, structural-key, and MACCS keys.
Fingerprints for screening: circular (ECFP and FCFP), path, topological-torsion and structural-key fingerprints, including the 166 MACCS keys.
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Offers optional GPU acceleration via separate packages for NVIDIA CUDA and Apple Metal, automatically used for large corpora.
Optional GPU screening is provided separately by smsd-cuda for NVIDIA GPUs and smsd-metal for Apple silicon Macs.
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Results include completion status, optimality proof, stopping reasons, and atom mapping indices for precise structural correspondence.
Every result reports completion, proved optimality and the reason it stopped, and maps atoms by their zero-based indices in the input molecules.
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Includes adapters for data-frame and machine-learning libraries like NumPy, pandas, PyTorch, and scikit-learn.
Python adapters for data-frame and machine-learning libraries, loaded only when you use them.
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Provides an MCP server with local molecular tools designed for integration with AI assistants.
An MCP server with local molecular tools for AI assistants, plus a container recipe.
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Supports packed fingerprint corpora with clustering capabilities for efficient large-scale screening.
Packed fingerprint corpora with clustering.
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Includes utilities for R-group decomposition, file input/output (MOL, SDF), SVG depiction, and molecular standardization.
Utilities, including R-group decomposition, MOL and SDF input, SVG depiction (C++ and Python) and molecular standardisation (Java).
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Substructure search with molecule or SMARTS queries, supporting non-induced or induced embeddings.
Substructure search with molecule or SMARTS queries, as non-induced (default) or induced embeddings
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Maximum Common Substructure (MCS) search for MCIS or MCES, connected or as disconnected fragments, with optional atom weights.
MCS search for the maximum common induced subgraph (MCIS) or the maximum common edge subgraph (MCES), connected or as disconnected fragments
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Fingerprint screening using circular (ECFP/FCFP), path, topological-torsion, and structural-key fingerprints, including 166 MACCS keys.
Fingerprints for screening: circular (ECFP and FCFP), path, topological-torsion and structural-key fingerprints, including the 166 MACCS keys
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Optional GPU screening for NVIDIA CUDA and Apple silicon Metal to accelerate fingerprint pre-screening.
Optional GPU screening is provided separately by smsd-cuda for NVIDIA GPUs and smsd-metal for Apple silicon Macs
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Includes an MCP server with local molecular tools for AI assistants and a container recipe.
An MCP server with local molecular tools for AI assistants, plus a container recipe
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Python adapters for data-frame and machine-learning libraries like NumPy, Pandas, and PyTorch.
Python adapters for data-frame and machine-learning libraries, loaded only when you use them
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Results report completion status, proved optimality, stopping reason, and atom mappings by zero-based indices.
Every result reports completion, proved optimality and the reason it stopped, and maps atoms by their zero-based indices
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Supports prepared queries for searching against many targets in turn or on multiple threads.
Prepared queries that can be searched against many targets, in turn or on several threads
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Limitation: Exact graph matching operations are restricted to CPU execution, even when GPU screening is enabled.
Exact graph matching always uses the CPU.
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Limitation: Packed fingerprint indexes do not utilize Apple Metal GPU acceleration.
Packed fingerprint indexes do not use Metal.
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Limitation: Native Windows CUDA support is unavailable; users must use Linux packages or WSL2 for GPU acceleration on Windows.
Native Windows CUDA is not provided. On Windows, use the Linux CUDA package or this image under WSL2.
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Limitation: Time-limited searches may return valid but unproven mappings; users must verify completion and optimality flags.
A search that reaches its timeout can still return a valid mapping, so check the completion and optimality flags before treating it as a proved optimum.
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Limitation: Java CLI and CPU Docker images bundle CDK under the LGPL-2.1 license, which may impose redistribution restrictions.
The Java CLI JAR with dependencies and the CPU Docker images bundle CDK under the LGPL-2.1.
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Limitation: Platform support is limited to CPU wheels, optional CUDA/Metal packages, and Java JARs; other platforms may require source builds.
The supported-platforms table covers the CPU wheels, the optional CUDA and Metal packages and the Java JARs.
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Limitation: Native Windows CUDA is not provided; users must use the Linux CUDA package or a Docker image under WSL2.
Native Windows CUDA is not provided. On Windows, use the Linux CUDA package or this image under WSL2
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Limitation: Searches that reach a timeout may return valid mappings that are not proved optima, requiring users to check completion and optimality flags.
A search that reaches its timeout can still return a valid mapping, so check the completion and optimality flags before treating it as a proved optimum
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
other · data · developer-tools