|
Algorithm Builder (AB) is a software system for
building QSAR, QSPR and SAR models and converting these to custom (in-house)
predictive algorithms. It allows the use of focused, company-specific
libraries to derive highly accurate models for the calculation of physical
properties and screening of biological activities for new compounds.
AB's unified interface of integrated modules replaces the "patchwork"
solution consisting of applications such as spreadsheets, statistics
programs, fragmental/fingerprint toolkits, and programming platforms
from multiple vendors.
The much anticipated solution that:
- Eliminates labor-intensive "trial-and-error" approaches
- Converts your proprietary data into predictive algorithms and filters
- Delivers workable solutions by combining methodologies from computational
and medicinal sciences
- ADME descriptors (optional modules): pKa, LogP, LogD, permeability, solubility
- Industry-leading molecular fragmentation and substructure matching algorithms
- Clean separation of calculation steps
- You can analyze your proprietary compound collections as well as published data to derive accurate, robust and easily editable models for decision support in compound profiling, screening, and lead optimization.
- The system is based on a graphical interface, centered around chemical structures, and does not require the user to do any scripting or programming.
- Combine fragmental, descriptor and similarity-based methods
- Use logical expressions, conditions and equations based on descriptors,
subfragments, internal interactions or any other chemical criteria
- Define additional external parameters in the algorithm - experimental data or third party descriptors can be incorporated
- Combine multiple sub-algorithms into general algorithms
|
- Over existing company interfaces via the Algorithm Launching Pack
(ALP)
- Using the Pharma Algorithms computational environment (including QSAR Builder)
The following algorithms were developed with Algorithm Builder and are
available commercially:
- LogP based on >9,000 structures
- Aqueous Solubility (N>5,000)
- Human Intestinal Absorption (N>1,000)
- Acute Toxicity LD50 (N>70,000)
- P-gp Substrate Specificity (N>1,000)
- Compound Solubility in DMSO (N>20,000)
Japertas, P., Didziapetris, R., Petrauskas, A. Fragmental Methods
in the Design of New Compounds. Applications of the Advanced Algorithm
Builder.
Quantitative Structure-Activity Relationships 21, 23-37 (2002).
Wiley Interscience subscribers can view the full
text of the article on-line.
|
