Home Tianwu Tech Completes Over RMB 100 Million Series A Funding to Accelerate Iteration of General AI for Protein Design

Tianwu Tech Completes Over RMB 100 Million Series A Funding to Accelerate Iteration of General AI for Protein Design

Nov 15, 2024 08:00 CST Updated 08:00
Matwings Technology

AI Protein Design Service Provider

36Kr has learned that Matwings Technology has recently completed an A-round financing of over 100 million yuan. This round of financing was led by Qiming Venture Partners, with participation from investors including Yonghua Investment, Jingtao Capital, and existing shareholder JoinCapital. The funds raised will be mainly used for the development of self-research projects, as well as the technical iteration and upgrade of AccelProtein™, a general artificial intelligence for protein design (originally developed by the Hong Liang team at Shanghai Jiao Tong University as part of the Pro series artificial intelligence).

Matwings Technology was founded in 2021, focusing on using AI for protein design, serving fields such as innovative drugs, in vitro diagnostics, and nutritional health. Core members of the founding team come from Shanghai Jiao Tong University and have been deeply engaged in the AI protein field for many years.36Kr previously reported

For a long time, proteins have played a crucial role in the biomanufacturing industry, especially those with excellent functions, which are vital for driving industry development. However, traditional methods of obtaining high-performance proteins often face challenges such as high costs, complicated processes, and low success rates.

Matwings Technology has independently developed a general artificial intelligence for protein design, AccelProtein™, which can directly predict protein functions. It breaks through the limitations of structure prediction methods represented by Alphafold in the industry, which "can only predict structures but cannot design functions," providing a new technical means for the efficient acquisition of high-performance proteins.

Specifically, AccelProtein™ leverages deep learning technology, utilizing a self-constructed dataset of 780 million protein sequences (including 500 million private data entries) and nearly 100 million functional labels, to understand the complex relationships between protein sequences, structures, and their functions. It achieves predictions from sequence to function and possesses the ability to efficiently predict protein functions with zero or minimal samples.

"Matwings Technology has expanded its business from its initial focus on protein design in the biopharmaceutical field to the field of synthetic biology, forming a development strategy that advances both 'biopharmaceuticals + synthetic biology.' It has achieved directed evolution and multi-objective optimization of proteins in various scenarios of biopharmaceuticals and synthetic biology, such as improving the thermal stability, activity, and affinity of proteins," Matwings Technology told 36Kr.

In the biopharmaceutical field, AccelProtein™ can play a role in multiple stages such as CMC, early screening, and candidate molecule optimization. Taking the CMC stage as an example, this artificial intelligence can enhance the activity, specificity, thermal stability, and pH stability of key tool enzymes and tool proteins. In the field of synthetic biology, relying on AccelProtein™ for enzyme modification, enzyme mining, and enzymatic process development can address the precise synthesis challenges of high-value molecules that are difficult to solve through chemical synthesis or natural extraction, providing efficient and green enzymatic solutions for pharmaceutical intermediates, flavors, health product ingredients, and cosmetic raw materials.

According to reports, since its establishment more than three years ago, Matwings Technology has successfully delivered over 30 protein design projects, covering enterprises in various fields such as innovative drugs, in vitro diagnostics, pharmaceutical intermediates, and nutritional health.