ACADEMIA

Software 4 Science Developments S. L., a Universidade de Santiago de Compostela spin-off, has developed AFFINImeter

Successful design of new and efficient products in a chemical lab is extremely complex. For instance, in order to create a new drug, the interactions between pathological proteins connected to diseases and the ligands capable to neutralize them (the potential active principle of the future pharmaceutical product) require to be studied in detail. Typically, scientists work with very complex chemical structures where the drug molecule cannot be chemically isolated or where proteins have several binding sites, each with a different affinity to the drug molecule. Every single factor must be studied both individually and in combination with each other.

The same argument applies to a wide variety of products having a great impact in our daily life: cosmetics, food products, new materials with specific physical or chemical properties, paints, solvents, molecular sensors for new technologies, industrial or household cleaning products, etc. Finding the most efficient protocol to solve a specific problem implies an arduous research process that is costly in time, human, chemical and economic resources.

Software 4 Science Developments (S4SD) is a start up company which takes care of laboratory needs through AFFINImeter, a modelling and analysis software that uses CESGA's computational resources. S4SD was created by members of the Soft Matter & Molecular Biophysics Group from the University of Santiago de Compostela (USC). Its flexibility and the multiple and powerful features of AFFINImeter make S4SD a ground-breaking company.

Ángel Piñeiro, S4SD co-founder and chief scientific officer, explains: “AFFINImeter is a computational platform for the analysis of molecular binding, aggregation and dissociation experiments. It is specifically designed to study complex systems. In order to do so, the researcher must first develop a model, which requires several phases: writing down every chemical reaction, translating them into equations, creating a computer code to solve these equations, and validating it before conducting the analysis. This process often requires years and, in many cases, it has been the main subject of PhD projects. With us, the researcher only has to worry about the first step, writing down chemical reactions; AFFINImeter will take care of the rest.”

The strong suit of the platform is that it complements the functionalities of the software installed by default in the laboratory equipments. Typically, such software is only able to offer up to 10 different simple models, thus limiting the versatility of the instruments. But AFFINImeter can offer “an unlimited number of analysis schemes, through a graphical building tool with which the user can create personalized models by defining the corresponding chemical reactions. The flexibility of our solution is unique. This allows us to exploit the full potential of experimental instruments. The obtained results are fully reliable and there is no room for calculation errors affecting the experiments conclusions.”

At present, AFFINImeter includes three applications. AFFINImeter ITC Simulator is an assistant to design ITC (Isothermal Titration Calorimetry) experiments. These experiments provide information on molecular interactions through the heat exchange involved in the studied processes. With this application, the researcher suggests different experimental conditions, and the simulator predicts the heat that will be obtained for each of them. In this way, the number of real experiments that the researcher has to carry out is reduced, thus optimizing the whole process.

AFFINImeter ITC Analysis has been designed to analyze actual experiments. The lab instrument gathers all the information from the experiment and the S4SD software, using CESGA's calculation resources, provides a complete characterization of the molecular interaction. Taking up again the example of the pharmaceutical industry, this application will rank the potential candidates to become an active principle as a function of their ability to bind a given protein.

AFFINImeter Spectroscopy, still under development, will analyze the results of several spectroscopic techniques that measure physical properties other than heat (e.g. absorption or emmision intensity of light, chemical displacements, etc). This software provides a complement or even an alternative to ITC to study physicochemical processes. Studying the same process by several independent methods will increase the reliability of the obtained results. 

Eva Muñoz, Surface Plasmon Resonance (SPR) expert, joined the team in January 2014 to collaborate on the implementation of this methodology in AFFINImeter Spectroscopy. She evaluates the tool both as a developer and a user: “I have been working in molecular interactions for years, focusing on systems that are hard to analyze. I found there has been a lack of tools that can help us to understand these complex systems which are really common in nature. As an user, I consider this a really powerful tool.”

CESGA's involvement in the project was really active. Carmen Cotelo, Senior Applications Technician at the Center, explains that there was an initial technical support stage focused on three aspects: the optimization of the code for the scientific application that S4SD was developing, in order to take better advantage of its potential; the creation of a web interface that is user friendly, attractive, easy to use and that could offer the results in visual and textual formats, both recognizable and useful; and the implementation of a connector to link the code to the web interface. Now CESGA provides its computational resources to the project. Its computational power is essential to run the supercomputing operations, which can take several hours in certain cases.

A fully functional beta version of AFFINImeter has been launched on July 8th of this year. The development team will continuously work to implement new functionalities. “Once the software to study molecular aggregation, dissociation and association processes is complete, we aim to include models to characterize other properties of the same systems, such as the kinetic description of the different reactions or changes in the molecular structure of the involved molecules”, continues Ángel Piñeiro. “Our long-term goal is to explore what other needs for scientific data analysis the market demands, and to implement them in the platform.”