Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • ADVERTISEMENT FEATURE Advertiser retains sole responsibility for the content of this article

Harnessing the power of quantum computing for drug discovery

Traditional computer-aided drug design (CADD) helps researchers screen for potential new drug-like molecules and predict their biological activity before completing laboratory experiments. However, for most systems, it takes too long to explore all possible compounds with drug-like characteristics, so searches are constrained to a relatively small pool.

CADD is an iterative process requiring a series of optimizations. “Once you find a molecule which binds to your target, you optimize it to bind better, be more soluble, be more stable, less toxic, etc.” Shahar Keinan, CEO of Polaris Quantum Biotech (POLARISqb), explained. “By starting from a much larger chemical space and transforming drug design to an optimization problem, we can find molecules with all the requisite properties in a single step”.

Keinan and Bill Shipman, Chief Technology Officer of POLARISqb, have been working towards this objective since co-founding the company in 2020. “With our quantum-inspired technology, we can efficiently scan large chemical spaces for potential drugs at unparalleled speed,” Keinan said.

POLARISqb built the first drug discovery platform that uses a quantum annealer, a specialized type of quantum computer that solves combinatorial optimization problems. By leveraging the principles of quantum physics, a quantum annealer can quickly solve very complex multi-object optimization problems that are prohibitively difficult for classical computers.

“Rather than using a brute-force approach to calculate all possibilities, the wave function of a quantum system scans all possibilities to find the best molecule out of billions the first time,” said Keinan.

POLARISqb specializes in translating chemistry into the language of quantum computers. “We know how to create chemical space in the format of quadratic unconstrained binary optimization (QUBO) algorithms1 that run on a quantum annealer for optimization,” she said.

A validated approach

POLARISqb is working on over 15 drug discovery projects and building a promising portfolio of lead molecules.

To demonstrate the validity of the approach, computational chemists at POLARISqb identified drug candidates that target the RNA-dependent RNA polymerase of the Dengue virus, a mosquito-borne viral disease that currently has no cure and threatens up to 40% of the world’s population2. The team built a chemical library of 1.3 billion compounds and used a quantum annealer to assess the library and to design molecular leads. Within six months they were able to prioritize molecules for experimental testing. Out of 30 final lead molecules, 10 shared similar motifs with molecules that are active in cells identified after seven years of effort by Novartis3.

Working with POLARISqb

POLARISqb has partnerships with several companies seeking to integrate quantum technologies and advanced artificial intelligence (AI) to discover novel lead compounds for a wide range of medical indications.

POLARISqb is working with Auransa to find therapeutics for diseases that disproportionately affect women. By combining their expertise in AI-driven biology and quantum-computing-based chemistry, they are searching and testing molecules that target a dysregulated pathway in triple-negative breast tumors.

“We are delighted with the progress we are making with POLARISqb and are continuing to tackle complex women’s diseases and bring much-needed therapeutics into the clinic,” said Pek Lum, CEO of Auransa.

POLARISqb is a service provider to companies developing small drug-like molecules against specific targets. POLARISqb builds and searches large chemical spaces based on the client’s criteria. “We are often approached to find drug-like molecules against difficult targets, such as G protein-coupled receptors (GPCRs) or protein–protein interactions (PPIs),” Keinan said. “We are very good at finding molecules that are hydrophobic enough to bind pockets in PPIs but soluble enough to reach them.”

POLARISqb can recommend ~100 lead molecules with the desired biological activity and properties, including synthesizability, within 3 to 4 months.

At the end of June 2023, POLARISqb will also be offering a Software-as-a-Service (SaaS) platform called QuADD that is executed on a quantum annealing system. Customers will be able to submit their target and small-molecule requirements to the QuADD platform, which will scan across a large chemical space while maintaining data confidentiality (Fig. 1).

Diagram showing how the QuADD platform identifies a top enriched library of candidates

Fig. 1 | The QuADD platform. Identifying a top enriched library of candidates in days.

Out of many billions of molecules, POLARISqb identifies the top enriched library of candidates in a matter of days. Customers can load these results into their proprietary CADD workflow, use their internal algorithms to further optimize the QuADD results, and speed up the hit-to-lead process. Proof-of-concept testing established that QuADD successfully reverse-engineered known drugs. For example, starting from the human P38 mitogen-activated protein (MAP) kinase protein-binding pocket, QuADD suggested a set of fragments that corresponded to Nexavar (sorafenib), a commercial oncology drug.

“To my knowledge, this is one of the first SaaS quantum-based products available on the global market and the only one for drug design,” said Keinan. “It’s increasingly difficult to find new lead molecules, but using QuADD to create novel, enriched molecular libraries of lead molecules will close this discovery gap.


  1. Maurice Benson, B. et al. Preprint at arXiv (2023).

  2. Whitehorn, J. & Simmons, C. P. Vaccine 29, 7221–7228 (2011).

    Article  PubMed  Google Scholar 

  3. Snelling, D. et al. Preprint at ChemRxiv (2020).

Download references


Quick links