Protein Structure Prediction and Design Service Platform Provider
The COVID-19 pandemic has not yet subsided, and monkeypox has already raised an "alarm".
On May 29 local time, the World Health Organization (WHO) issued a disease information update, assessing the global public health risk of monkeypox as moderate. According to the WHO update, from May 13 to 26, 23 non-endemic countries and regions reported 257 confirmed monkeypox cases to the WHO, with around 120 suspected cases.
In response to potential epidemic risks, a group of biopharmaceutical-related institutions has begun to take action.
Recently, Zelixir, a bio-computation platform company, announced that it has released protein structure predictions for over 600 proteins in the monkeypox proteome, along with related protein function annotations. All prediction results were achieved through the company's self-developed full-stack protein computation online platform, YuCloud ZCloud, which is expected to assist scientists worldwide in vaccine and drug design for the monkeypox virus based on protein structures.
"We hope to accelerate the progress of monkeypox vaccine and drug development through our work. As the research advances, we will continuously update and release the latest findings," said Dr. Wang Sheng, CEO of Zelixir.
Public information shows that monkeypox is a rare viral zoonosis (a virus transmitted from animals to humans). The clinical symptoms observed in patients are similar to those seen in smallpox patients, but with a milder clinical severity. Following the eradication of smallpox in 1980 and the subsequent cessation of smallpox vaccination, monkeypox has become the most serious poxvirus.
In the recent prediction process of monkeypox-related protein structures, Zelixir selected the complete genome sequence (ID: ON563414.2) of the monkeypox epidemic strain from May 2022 (Reference 1), the entire proteome of the 2018 West African (WA) virus strain (Monkeypox virus isolate MPXV-UK\P3, ID: MT903345.1) (Reference 2), and the entire proteome of the 1996 Congo Basin (CB) virus strain (Monkeypox virus strain Zaire-96-I-16, ID: NC_003310) (Reference 3).
With the help of ZCloud, which is equipped with the AlphaFold2-Batch algorithm, Zelixir quickly achieved structural predictions for all proteins of the aforementioned strains, generating over 600 prediction results. It is reported that AlphaFold2-Batch is a novel algorithm developed by Zelixir based on AlphaFold2, achieving nearly two orders of magnitude of heterogeneous acceleration compared to the original algorithm, enabling rapid prediction of protein structures related to the monkeypox virus.
In addition, for the three-dimensional structure of each protein, Zelixir has also used the PointSite algorithm, which was primarily developed by the team, to conduct high-precision prediction and annotation of small molecule binding regions. The prediction results published this time include functional annotations and analyses of the proteomes of the 2018 West African strain and the 1996 Congo Basin strain.

Currently, Zelixir has chosen to make all prediction information and functional prediction results publicly available, and scientists worldwide can access the relevant information on the company's official website (https://www.zelixir.com/Monkeypox/index.html).
In the past week, Zelixir has actively explored and established a Monkeypox virus prediction model to accurately predict the prevalence of Monkeypox. Based on the latest situation of Monkeypox, Zelixir has updated the Monkeypox prediction model and continuously optimized the global Monkeypox prediction system.
In the view of industry insiders, although these are only approximate protein structures predicted by AI calculations, under the background of the global spread of monkeypox virus, these protein structures may all contain effective sites for new drugs or therapeutic agents. They have very important guiding significance for subsequent drug and vaccine development and can make practical contributions to curbing the monkeypox virus.
In fact, the prediction and design of protein structures are precisely one of Zelixir's "core competencies."
Public information shows that Zelixir was founded in April 2021. It focuses on protein structure prediction and design technology, applying it to target discovery, drug design, enzyme engineering, biosynthesis catalysis, and other fields. It provides auxiliary services for drug research and development as well as R&D and production services for synthetic biology products.
Currently, Zelixir has independently built a complete system for protein structure prediction, design, and production. Specifically, the company has developed Yuyun ZCloud, a protein structure prediction and R&D support platform; created Zhishan ZPod, an all-in-one engineering solution for protein structure computation; and established Zhihui ZBot, a high-throughput automated wet lab platform, significantly improving the success rate of drug development. This has formed a high-tech enterprise integrating infrastructure, protein design services, and synthetic biology products into a complete industrial chain. Based on these achievements, within less than a year of its establishment, Zelixir consecutively secured angel round funding led by CDH Investments and Langyu Investment, as well as Pre-A round funding led by Hillhouse Ventures.
In the field of drug research and development, Zelixir, leveraging big data and AI for protein structure prediction as well as protein-ligand binding energy calculation and optimization, can significantly enhance the efficiency of the drug discovery phase. To date, the company has completed over 300,000 protein structure predictions and provided R&D support services to nearly 30 clients.
In the future, Zelixir's cloud computing service platform will also introduce new drug research and development services such as fully automatic high-throughput virtual screening, structure-based small molecule generation, fully automatic high-performance free energy perturbation, and small molecule property prediction.
"Each module represents a specific step in the biotech research process. We hope that through our self-developed service platform, we can link up each step to help biopharmaceutical companies complete protein-related research for new drug development at the touch of a button. Afterwards, pharmaceutical companies can smoothly complete drug discovery by simply referencing the computational results," said Dr. Wang Sheng.
While continuing to provide protein prediction and design services during the new drug development process for biopharmaceutical companies, and solidifying its position as a CRO company in the computational field, Zelixir will also expand into synthetic biology based on its existing technological advantages.
It is reported that, through a pure AI computing and design approach, Zelixir can modify and design enzymes for pharmaceutical intermediates from natural substrates within a short period. Data shows that the catalytic activity of the modified enzymes increased approximately 5-7 times, and their thermal stability improved by nearly 30 degrees. At the same time, it is possible to graft and modify the enzyme’s catalytic pocket, creating a previously non-existent pathway to develop products through purely biological methods. "This is a highly versatile technology that can essentially be applied to all enzymes and proteins. Compared to traditional methods, it is more precise, faster, and less costly."
"We hope to understand life through computation, improve life through synthesis, and reconstruct life through design, becoming a world-leading provider of synthetic biology technology and products, bringing a visible revolution to biotechnology," said Dr. Wang Sheng.
References:
[1]https://virological.org/t/first-draft-genome-sequence-of-monkeypox-virus-associated-with-the-suspected-multi-country-outbreak-may-2022-confirmed-case-in-portugal/799
[2] http://exon.gatech.edu/genemark/genemarks.cgi
[3]https://www.uniprot.org/uniprot/?query=monkeypox%20virus&fil=organism%3A%22Monkeypox+virus+%28strain+Zaire-96-I-16%29+%28MPX%29+%5B619591%5D%22&sort=score
Leifeng.com(Official Account: Leifeng.com)
Leifeng.com copyrighted article, unauthorized reproduction is prohibited. For details, seeReprint Notice。