Nuojing Technology is seeking 30 million yuan in angel round financing.
AI-designed oral cyclic peptides target "undruggable" targets, Nuojing Technology seeks angel round financing
New drug R&D is facing a structural contradiction: on one hand, targets with large and shallow flat interfaces such as protein-protein interaction (PPI) have long been classified as "undruggable", as they exceed the theoretical action coverage area of small-molecule compounds; on the other hand, antibody drugs that can hit such interfaces with high affinity are limited by their molecular weight and difficult to penetrate cell membranes, making them unable to reach a large number of intracellular targets. According to the generally accepted statistical standard in the industry, a new drug takes an average of more than ten years from discovery to marketing, with a cumulative investment of more than one billion US dollars, and the overall success rate is less than 10%. The target verification and lead compound optimization stages consume the most cycles and experimental resources.
Shenzhen Nuojing Technology Co., Ltd. chooses synthetic cyclic peptides as its entry point. The molecular weight of cyclic peptides is usually in the range of 1 to 2 kilodaltons, between small molecules and antibodies, which not only has the ability to cover a large binding interface, but also retains the possibility of transmembrane and tissue penetration. Huang Qibai, founder and CEO of the company, said that Nuojing Technology hopes to turn the molecular form of cyclic peptides, which is "theoretically feasible but extremely difficult in engineering", into R&D materials that can be stably produced by computational models, instead of relying on high-throughput screening by luck.
From molecular generation to crystallization prediction, the platform covers multiple links in drug discovery
The core product of Nuojing Technology is the OriMed® intelligent drug R&D platform. The platform has three main modules, corresponding to different links in the drug discovery chain respectively: OriGen™ is responsible for the design and generation of molecules such as cyclic peptides, OriOmics™ is responsible for target discovery based on multi-omics data, and OriTal™ is oriented to crystallization prediction in the solid-state R&D link. In addition, the company is also equipped with automated high-throughput experimental capabilities to complete the wet experimental verification of calculation results.
In the molecular generation stage, OriGen™ works not by screening from existing compound libraries, but by directly generating brand-new sequences. Its modeling idea includes three levels: first, to build a dynamic system close to the real physiological environment of the human body, rather than making predictions under isolated in vitro conditions; second, to describe molecules at multi-level and multi-scale, taking into account atomic fragments, amino acid sequences and physicochemical properties at the same time; third, to integrate multiple factors affecting peptide druggability such as affinity, membrane permeability and stability into the same round of multi-objective optimization, instead of optimizing the activity first and then conducting other property tests one by one. The molecules output by the model can contain non-natural amino acids, and are supported by a synthetic experience library to reduce the risk of "calculable but unmanufacturable".
The test data provided by the company shows that among the generation results of OriGen™ on targets such as PD-L1, PDGFR, IL-7Rα, IL-4Rα, the highest binding interface credibility (ipTM) reaches 0.894, the highest single structure prediction confidence (pLDDT) is 95.60, and the number of hydrogen bonds in some sequences is 4. The model also produces comparable results on new targets that are not included in the training dataset. The company said that its design process has a single cycle of 24 hours, the wet experiment verification cycle is about 3 weeks, and the hit rate reaches 80% (taking less than 10 nanomolar as the threshold), that is, about 20 sequences tested can obtain candidate molecules with further development potential.
It should be pointed out that the above data comes from tests organized by the company itself, and has not been published in peer-reviewed literature or third-party blind tests. In the AI pharmaceutical field, there is a common gap between model indicators and actual binding activity, and such data is more suitable as evidence of platform capabilities rather than a conclusion of druggability. The real test still comes from wet experiments and subsequent development links.
In the target discovery stage, OriOmics™ targets biotechnology companies engaged in ADC, PROTAC and TCE in the oncology field. The input data types include Bulk RNA-seq, single-cell RNA-seq, TMT/DIA/PRM proteome, phosphorylated proteome, genomic data such as mutation and copy number, metabolite profiles, and epigenetic data such as methylation and ATAC-seq, and integrates public databases such as TCGA, GTEx, HPA and clinical literature evidence. Its process starts from differential signal mining, goes through tumor-enriched target screening, normal tissue safety window prediction, protein-level evidence verification, and then uses single-cell and spatial omics data to confirm the cell specificity and tissue distribution of targets. The company stated that this module has cumulatively produced more than 30 potential new targets.
Prioritize service revenue first, then launch self-developed pipelines
In terms of commercialization path, Nuojing Technology has adopted a dual-line model of "platform services first, self-developed pipelines follow up". The platform side delivers in the form of custom R&D services, targeting innovative pharmaceutical companies and biotechnology companies. The process includes demand docking, scheme design and technical implementation meetings, experimental verification and report delivery. The company disclosed that its molecular generation model, target discovery model and crystallization model have all been launched and provided services to customers, leading pharmaceutical companies have made purchases, and some drug discovery orders have been delivered; partners include Fuxing Pharmaceutical, Livzon Pharmaceutical, and a joint AI pharmaceutical R&D center has been built with Livzon Pharmaceutical. In addition, the company has reached cooperation with Huke Biotechnology in the fields of immunotherapy and cytokine drugs, involving the joint construction of a data system for large-scale cell preparation and model research of cytokine combinations.
In terms of self-developed pipelines, the company placed its first clearly announced project on the IL-4Rα target, with the indication of atopic dermatitis and the molecular type of oral cyclic peptide, which is currently in the lead compound optimization stage. The background of this choice is: the mechanism of the IL-4Rα pathway has been verified by antibody drugs, but existing therapies are mainly administered by injection, while atopic dermatitis is a chronic disease that requires long-term medication, and oral regimens have obvious room for improvement in compliance. Among the candidate molecular data provided by the company, the top-ranked sequence has an ipTM of 0.940, a pLDDT of 89.68, a shielded surface area (ΔSASA) of 248.9 square angstroms, a hydrogen bond number of 2, an interface residue number of 9, a radius of gyration of 5.13 angstroms, and a contact number of 60.
The expansion of the market space provides external conditions for such attempts. The cyclic peptide track has seen intensive commercial progress in 2026: in March, the oral IL-23R cyclic peptide co-developed by Johnson & Johnson and Protagonist Therapeutics was approved by the FDA for moderate to severe plaque psoriasis; in July, Merck's oral PCSK9 cyclic peptide Enlicitide was approved by the FDA, becoming the first oral PCSK9 inhibitor, with LDL-C reductions of 56% and 59% at week 24 in the pivotal phase III clinical trial; in August, another cyclic peptide Rusfertide from Protagonist was also approved. According to statistics from industry research institutions, the number of global cyclic peptide R&D pipelines has increased from 72 in 2021 to 93 in 2026, and the total value of global cyclic peptide licensing transactions in the first three quarters of 2026 reached 6.17 billion US dollars. Among domestic enterprises, Yuansheng Peptide completed a $150 million Series B financing in March 2026, Zhongsheng Quantopeptide completed a Series C financing of over 300 million yuan in May 2026 and launched IPO counseling on the Science and Technology Innovation Board, and Salubris's oral PCSK9 cyclic peptide was approved for clinical trials by NMPA in July 2026.
However, the warming up of the track also means that the competition threshold is rising rapidly. Domestic enterprises in the same cyclic peptide design track as Nuojing Technology have covered multiple technical routes: Yuansheng Peptide's Synova™ platform adopts a high-throughput screening paradigm, Zhongsheng Quantopeptide relies on a peptide entity library of nearly 500 million magnitudes and PICT information compression technology, while Jingtai Technology covers the complete chain from Hit to PCC and is equipped with automated synthesis and high-throughput library construction. In contrast, Nuojing Technology is still at an early stage in the advancement of its self-developed pipelines, and whether its platform advantages can be transformed into customer repurchases and pipeline value still needs more verified delivery cases to support.
Cross-border industry-university-research team, the round plans to raise 30 million yuan
Nuojing Technology was founded in Shenzhen, with offices and R&D centers in Shenzhen, Changsha and Boston. It is positioned as an AI-driven innovative drug R&D service platform. Huang Qibai, founder and CEO, is a serial entrepreneur who once worked at ByteDance and Xiaomi Group, responsible for large-scale complex system architecture and team management. Xia Zhinan, Chief Scientist, received his bachelor's and master's degrees from China Pharmaceutical University, obtained his doctorate from the University of Kentucky, and then engaged in postdoctoral research at Harvard Medical School, Brigham and Women's Hospital and Boston Children's Hospital. He holds 25 patents and 30 SCI papers, and co-founded the Chinese Antibody Society; his professional experience includes Wyeth, Pfizer, Synageva and Moderna. He once led the half-life extension research of multiple novel biological entities and obtained multiple IND approvals. After 2018, he founded Abimmune Bio, and from 2020 to 2023, he founded DynamiCure Biotech and promoted it to the clinical stage before exiting as a whole.
In terms of the technical team, She Ziyu, CTO, is a doctor of computational drug discovery from the University of Basel, whose research directions cover macrocyclic peptide and non-natural amino acid peptide generation, protein-ligand interaction modeling and free energy prediction, with a total of about 350 citations; algorithm expert Florian Hinz graduated from ETH Zurich, with research accumulation in equivariant neural networks, normalized flows, GFlowNet, diffusion models and other directions; computational scientist Qian Yumin has long been engaged in computational chemistry and molecular design research, and has published more than 70 papers in journals such as Nature sub-journal and JACS, with a total of more than 8000 citations. The pharmaceutical business is led by Zheng Zhengcai and Zhang Zhenyi. The former has 24 years of experience in peptide drug R&D and management, and has built a peptide CDMO technology platform in listed companies such as Asymchem. The latter is a doctor from Shanghai Jiao Tong University, who once served as a senior researcher at WuXi AppTec and won the WuXi AppTec-Merck and WuXi AppTec-GENENTECH Excellence Contribution Awards, and also served as the head of small molecule and protein crystallography in the Bruker SCD China region. Nancy Su, Head of Pharmaceutical R&D, has worked at AstraZeneca for 22 years, responsible for drug pipeline discovery and high-throughput screening.
In terms of financing, this round of the company is an angel round, planning to raise 30 million RMB, transferring 10% of the shares, and the expected capital use period is 12 months. According to public industrial and commercial information, Huang Qibai is the legal representative and direct controlling shareholder of Shenzhen Nuojing Technology Co., Ltd., with a shareholding ratio of about 73%. The company stated externally that the raised funds will be mainly used for platform model iteration, pipeline advancement and team expansion.
From the perspective of industry position, Nuojing Technology is facing both opportunities and pressure. The competition focus of AI pharmaceuticals has shifted from "whether the model can generate molecules" to "whether it can stably deliver druggable and verifiable candidate molecules", and the molecular form of cyclic peptides has just completed the commercial verification of druggability in 2026. For Nuojing Technology, the key to the next stage is not the model indicators themselves, but the number of delivery cases, customer repurchase rate, and whether the self-developed pipeline can be advanced to the IND-enabling stage as scheduled.