Galux claims ‘one-day 해시 게임 바카라 design’ as it enters the global top-three AI race

[by Ji, Yong Jun] "Artificial intelligence (AI) can now design antibodies in a single day. A paradigm shift is underway, fundamentally transforming the very process of drug development."

Park Tae-yong, Executive Vice President of Galux, made this remark during an interview with <THE BIO> conducted on November 27 at the company’s headquarters in Gwanak District, Seoul.

The traditional process of antibody ‘discovery’ is evolving into one of ‘design’ as it merges with AI. Whereas past antibody development relied on inducing immune responses in animals and screening vast libraries, a method often described as a ‘pray-meta’ approach, the convergence of biotechnology and AI is ushering in an era of highly sophisticated ‘engineering.’ Notably, last year's Nobel Prize in Chemistry was awarded to three scientists who advanced the application of AI in protein structure prediction and de novo protein design, a milestone that is accelerating the pace of AI-driven innovation in the antibody field.

Galux, a Korean AI-driven drug development company, is emerging at the forefront of this transformation. Over the past two weeks, the company, working in collaboration with U.S.-based Nabla Bio and Chai Discovery, has solidified its standing as a leader in the global pharmaceutical sector by presenting research findings that advance its ‘de novo’ antibody design platform, which generates antibody sequences entirely from scratch.

Galux's major accomplishments include reducing the 해시 게임 바카라 discovery timeline from more than a year to under a month and demonstrating that the differences between AI-predicted 해시 게임 바카라 structures and actual antigen-binding structures fall within the atomic-level error margins. Building on this precision-driven design capability, the company is now concentrating on improving the clinical suitability of diverse 해시 게임 바카라 modalities and preparing for the next phase aimed at increasing clinical success rates.

◇AI-driven design reduces 해시 게임 바카라 discovery timelines from a year to one month

Vice President Park Tae-yong expressed in the interview, "We have moved beyond the traditional approach of waiting for a suitable antibody to appear by chance and have now established a system capable of precisely designing antibodies with the desired properties from the outset. This shift is not merely a matter of improving efficiency; it marks a fundamental turning point in antibody development."

Galux recently reported research findings showing that its de novo antibody AI design platform, ‘GaluxDesign,’ has reduced the antibody discovery process, which typically requires more than a year, to under a month. As a result, the company achieved a 31.5% success rate with only 50 antibody designs, eliminating the need for extensive library screening. The platform’s precision was further demonstrated by narrowing the error in antibody-antigen binding structures to 1 angstrom (1Å, 10-10 meters).

Traditional antibody discovery methods relied largely on immunization, in which the target antigen was injected into animals to elicit an immune response, or on screening approaches that selected candidate binders from vast antibody libraries ranging from 10-6 to 10-9 variants. Even when a promising antibody was successfully identified, development timelines were prolonged by complex downstream steps such as humanization and property optimization, and the likelihood of securing a high-quality antibody with that precise target binding remained uncertain. In essence, conventional antibody discovery methods amounted to a ‘game of chance.’

Galux is now adopting an approach that ‘designs’ antibodies from the ground up using AI. "We input the antigen structure into the AI system ​​and design the antibody around a pre-defined binding site. As long as the binding occurs as intended, we can secure the antibody we are targeting," Park explained.

The most significant breakthrough is the reduction in development time. GaluxDesign requires just a single day to generate an antibody design, while protein synthesis, binding evaluation, and antibody testing take approximately one month. "At present, the process takes about a month, but with further optimization of our evaluation system, we expect to shorten it to under a week. Our goal is to reduce the timeline to one week within this year," Park remarked.

◇Achieves 31.5% success rate across 50 designs, confirming 1Å precision

Park noted that the most significant achievement of this study lies in the precision of the antibodies designed using GaluxDesign. Previously, creating a library required designing 10,000 to 100,000 antibodies and then screening them, a process that imposed substantial time and cost burdens. The integration of 해시 게임 바카라, however, has dramatically reduced this burden.

"This means that even when only 50 antibodies are designed, roughly 30% of them will actually bind to the intended target," Park explained. "This success rate is determined by observing that a signal appears for the desired target in an ELISA assay, while no signal is detected for other targets."

"When we superimposed the designed antibody structure with the actual structure confirmed through 'cryo-EM,' the deviation was approximately 1 angstrom. This corresponds to a level of precision comparable to the hydrogen-oxygen bond length or the carbon–carbon bond length in a water molecule," he emphasized.

ELISA is an analytical technique that measures antigen-해시 게임 바카라 binding signals optically to determine the presence and quantity of a specific protein. Cryo-EM, meanwhile, is a method that visualizes biomolecules at cryogenic temperatures using an electron microscope to analyze their structure at atomic resolution.

◇"Competing with two global leaders for the top spot in AI antibody design"

Only a small number of companies worldwide have demonstrated de novo antibody design technologies at the experimental stage. "Only about five companies globally have validated this technology through actual experiments," Park explained. Specifically, these include Nabla Bio, David Baker Group (Xaira Therapeutics line), Chai Discovery, Tencent, and Google DeepMind.

"Galux, Nabla Bio, and Chai Discovery are the only companies that have demonstrated antibody success rates using a relatively small number of designs. In terms of design precision, we can confidently say that we are among the top three globally," Park emphasized.

Park also underscored the potential value that AI-engineered antibodies can generate. "In the case of the contract Generate Biomedicines signed with Amgen, the upfront payment alone is around KRW 70 billion (approximately USD 47.6 million), and milestone payments total KRW 3 trillion for the discovery of five targets. Royalties following commercialization also reach double-digit percentages of revenue," he noted.

Park predicts that Galux will use its 해시 게임 바카라 design platform to secure deals on par with those of global pharmaceutical companies, covering the full spectrum from candidate discovery and optimization to out-licensing.

Building on this achievement, Galux is extending its 해시 게임 바카라 design research to encompass difficult-to-target proteins such as GPCRs and ion channels, which remain challenging for existing technologies. In addition, through collaborations with various Korean and international pharmaceutical companies, the company plans to accelerate the development of differentiated AI-designed antibodies to advance rapidly into clinical trials. Galux is also pursuing the development of various 해시 게임 바카라-based modalities, including multi-antibodies and 해시 게임 바카라-drug conjugates (ADCs).

Ultimately, Park's vision for AI-driven antibody design centers on ‘increasing clinical success rates.’ "While reducing time and costs is important, the real objective is to raise clinical success rates. Beginning with antibodies, the most natural therapeutic modality, we aim to build a platform capable of designing entirely new modalities 50 or even 100 years from now," he stated.