You can use experimentally measured logP values to train the algorithms in ACD/LogP. Improve prediction accuracy and make the model more relevant to your chemical space or project.

Both the Classic and GALAS algorithms can be trained even if you aren’t a computational chemist or software engineer. The Consensus model automatically uses training data applied to the underlying algorithms.

Classic Algorithm

Based on >12,000 experimental logP values, the Classic algorithm uses the principal of isolating carbons.

GALAS Algorithm

The GALAS model is Global, Adjusted Locally According to Similarity. The algorithm is based on a training set of >11,000 compounds and provides a value for logD that is adjusted with data from the most similar compounds.

Consensus Model

The consensus model uses both Classic and GALAS algorithms. Each algorithm is weighted more heavily in regions of chemical space in which it performs most reliably.

Predicted logP is likely to be more accurate than experimentally measured when:

• The logP value is outside the reliably measurable range (+8 to -3) obtained by traditional experiments.
• Tautomeric forms exist. In most cases, logP values cannot be measured for every tautomer.
• A compound contains acidic and basic groups (e.g., amino acids, peptides, nucleosides). Since these molecules exist in various ionic forms at any given pH, the concentration of the unionized form is negligible.

The partition constant (P) is a measure of how hydrophilic (‘water-loving’) or hydrophobic (‘water-fearing’) a neutral (uncharged) molecule is. It represents the tendency of a compound to differentially dissolve in these two immiscible phases (typically Octanol and Water). The partition coefficient is also referred to as K_ow and the octanol/water partition coefficient.

LogP prediction models estimate this value as a logarithmic ratio (logP, ClogP, or AlogP). The partition coefficient acts as a quantitative descriptor of the hydrophobicity of a compound.

Similar to logP (or clogP), logD is also a descriptor of hydrophobicity but it is not limited to describing the neutral molecule. LogD is a measure of the hydrophobicity for ionizable compounds which takes into account pH dependence.

Hydrophobicity (as determined by logP) can help explain or predict the behavior of a compound and is useful in many industries:

Pharmaceuticals—LogP helps medicinal chemists assess drug likeness. In pharmacokinetics it can help determine the ADME profile—the ability of a drug to be absorbed, successfully reach the intended target, be metabolized and excreted; and in pharmacodynamics to understand target receptor binding.

Agrochemicals—K_ow values are used to help develop herbicides and insecticides. Partitioning values help determine whether a compound will reach its intended action site and the likelihood of environmental pollution.

Environmental—Partition coefficients are used to model the migration of dissolved hydrophobic organics in soil and groundwater to help assess waterway pollution, and toxicity to animals and aquatic life.

Consumer Products—An understanding of partitioning is used in the formulation of cosmetics, dyes, household cleaners, and many other products.