Modifying wood properties with the aid of laccase makes it possible to develop waterproof or antimicrobial wood surfaces, for instance. The enzyme is able to alter the chemical structure of the wood on its surface without changing the structure of the material, explains the Swiss Federal Laboratories for Materials Science and Technology (Empa)
The problem is that there are many variants of laccase, not all of which react with the desired substrate. This makes it important to identify suitable laccase-substrate pairs. Conventional simulations require a high computing capacity and are very time-consuming and expensive, which is why Empa researcher Mark Schubert developed the “smart wood search engine”.
Schubert turned to deep learning, a type of artificial intelligence that trains a computer program to recognize patterns from large amounts of data from technical literature and experiments. For example, it can learn which laccase oxidizes which substrate. According to Empa, the search works even if not all details about the chemical mechanism are known.
Schubert is continuously training his search engine, even with partly unusable “data piles”, allowing it to “recognize what it can use and what not”.
The “smart wood search engine” is already being used by industry. For instance, Schubert is collaborating with the insulation board manufacturer Pavatex, which uses laccase in the production process.
More News
Meet with an expansion expert
Our services are free of charge and include:
- Introduction to key contacts in industry, academia, and government
- Advice on regulatory framework, taxes, labor, market, and setting up a company
- Custom-made fact-finding visits, including office and co-working space