The development of pharmaceutical is driven by two main goals: therapeutic advantages and time to market. Only 10% of drug molecules identified in research make it right through the long process of development. This means that many potential drugs do not make it to market, wasting time, effort and money. DFAS toolkits can help you to improve the efficiency of drug development.

DFAS is drug formulation assisting software. Achieving the optimal formulation of a drug candidate can determine the future development of that drug. Formulation is the process of mixing ingredients in such a way as to produce a new or improved product. In the drug industry, this usually starts with the selection of the most appropriate form of an Active Pharmaceutical Ingredient (API), and will include other compounds intended to make sure the API is still a standardized dose when taken by a patient. In the beginning, this may have been done using a trial-and-error approach, with a senior formulation scientist using the benefit of long experience to suggest what compounds would blend well together.

Siemyx develops software solutions to aid R&D center to overcome their research and devepompent challenges. DFAS provides organizations the capability to improve efficiency by delivery of time savings in bringing new products to market.The application is designed to assist in the process of developing new formulations. This complex task is critical to adding value in chemical, pharmaceutical, and consumer product businesses. Not only must a company produce materials with better properties, it must also understand regulatory constraints, customer requirements, cost implications, and manufacturing issues. These often conflict and must be balanced against each other.

Creating predictive models

The application's architecture puts no restrictions on the type of model that can be incorporated: linear regression, neural network, analytical - all of these types of model can be used. Three types of model builders are supported: a Genetic Function Approximation (GFA) algorithm, a Linear Regression algorithm, and a Bayesian Neural Net (BNN) algorithm. Each of these algorithms is used at a different point in the overall product optimization cycle..

Multiobjective optimization

Modern materials design requires that a wide range of performance criteria be met, thus, many objectives must be simultaneously satisfied. Finding the optimal amounts of each ingredient, or the optimal values of the processing parameters is a core skill of the practicing formulator. The application provides two methods for rationalizing and automating this task: weighted optimization and Pareto optimization.