Aaron Tech-Pro Private Limited

Home > Products > VisiQuest > Chemoinformatics Toolkit

Chemoinformatics Toolkit
Grouping and Clustering Optimized for High-Throughput Screening

AccuSoft has partnered with MESA Analytics & Computing to bring you a new, leading-edge toolset; the VisiQuest Chemoinformatics Toolkit. This toolkit enables you to quickly prototype and perform:

Similarity Searching
Use MDL MACCS Key Fingerprints and translate SMILES. Similarity Searching with the Tanimoto, Cosine, Euclidean, Hamman, or Tversky similarity or dissimilarity measures.
Clustering
Clustering large datasets, either non-overlapping or overlapping clusters, with the Tanimoto, Tversky, Euclidean and other measures.
Cluster and Active Compound Inspection
Use ChemTattoo(TM) to identify substructure commonalities among a group of compounds with key-based fingerprints.
and more

With the Chemoinformatics Toolkit, based on state-of-the-art MESA Analytics & Computing technology, users are given access to an additional set of powerful data mining solutions that cover a variety of early drug discovery data analysis techniques.

Chemoinformatics Toolkit Functions

MDL MACCS Key Fingerprints, 164 bit or 320 bit versions. Manipulate SMILES linear notation for database similarity searching, clustering, diversity, QSAR, & more.
Similarity Searching Measures: Tanimoto, Cosine, Euclidean, Hamman, Tversky, or composite similarity or dissimilarity measures.
Clustering measures output Ordered Pair, Sparse Matrix, or Square Matrix.
RussianDoll. Takes as input an asymmetric matrix and outputs a transformed asymmetric matrix that groups like sub-features. Useful in structural and shape clustering for lead hopping. This algorithm is unique to Mesa.
Additional Clustering allows for fast, space efficient clustering of very large data sets:
- Non-overlapping or overlapping symmetric clustering using Taylor-Butina algorithm with ambiguity measures
- Non-overlapping or overlapping asymmetric clustering using Taylor-Butina algorithm with ambiguity measures
- Non-overlapping or overlapping symmetric clustering using K-means with ambiguity measures
- Non-overlapping or overlapping symmetric clustering using K-modes with ambiguity measures
- Reciprocal Nearest Neighbor implementations of non-over lapping, symmetric Wards, Complete Link, and Group Average algorithms with ambiguity measures
- Non-overlapping or overlapping symmetric clustering using Jarvis-Patrick algorithm with ambiguity measures.
ChemTattoo (TM) identifies substructure commonalities among a group of compounds with key-based fingerprints. Generates the atom counting and coloring display information and produces a collection of important modal statistics. Visualization of the structures can be performed by numerous atom coloring depicters.
Chemical Diversity algorithm, a hybrid exclusion region algorithm that takes into account similarity measure distributions, useful in compound selection in building compound libraries.
MDS for cluster and diversity analysis visualization.
and much more!