It’s Like Nothing Else

Aria Pharmaceuticals
4 min readFeb 3, 2020


by: Andrew A. Radin

Today I received an email from an influencer in the life science community. Paraphrasing a bit, here’s what she said:

“We have reviewed nearly 400 companies in this space — so much going on! How are you differentiated from others?”

Rather than email her individually, I viewed this as the perfect opportunity to share how twoXAR Pharmaceuticals is different. Our work is like nothing else.

We are a pharmaceutical company that uses AI in the discovery of new medications

To begin to answer the question we must start with differentiating ourselves against the hundreds of companies using AI in life sciences. These companies cover everything from anatomy to zoology and everything in between. While there are parallels to our work in these areas, we differentiate ourselves from the broad spectrum of AI life science companies with our focus: We are a pharmaceutical company.

We create first-in-class small-molecule medications

New medications come in different physical forms such as small and large molecules. In some cases, new medications are enhancements upon standards of care. These so-called best-in-class medications improve properties such as potency, half-life or reduction of toxic side effects. Other times new medications work on new biology that has never been seen in a medicine in a disease area. These so-called first-in-class medications offer new solutions to unmet medical need. We differentiate ourselves from other pharmaceutical companies by what we produce: first-in-class small molecule medications.

We reduce the time from program initiation to lead compound by three years

To produce first-in-class medications, drug discovery and development experts have used a standard process for decades. Understanding pathogenesis, selecting targets, developing bioassays and subsequent high throughput screens are the typical first four steps of a first-in-class discovery effort. These stages can take over three years to complete. We replace these four steps with a process that runs completely in-silico that produces the same output (a pre-clinical ready molecule) in less than a month. We differentiate ourselves from other discovery processes by reducing the time from program initiation to in vivo results by three years.

We improve the success rate at in-vivo efficacy milestones by 30x

In first-in-class programs, others test about one hundred compounds at in vivo efficacy milestones with one or two molecules showing a promising signal. Part of the reason for this high false-positive rate is that even with modern computational approaches, reliance on homogeneous data sources doesn’t illuminate the broad spectrum of biological knowledge that required to eliminate false positives. We have created new AI algorithms that can handle disparate data sets allowing us to create a consensus view of disease and drug that significantly reduces false positives at in vivo milestones. Across our disease programs, we see on average 3 of 10 of our screening compounds with efficacy signals in vivo. We differentiate ourselves from other AI modeling methods by reducing the false positive rate 30x.

Our technology has been proven in over 18 disease areas

We’ve seen the claims from many in our space about how AI is going to transform drug discovery and development, but we’ve seen few working examples of technology producing results at high efficiency levels. Our company has initiated a wide range of programs in orthogonal disease areas, all starting with creating new models of pathogenesis to derive medications. We have 18 active development programs today in preclinical stages, a handful of which have been screened down to a lead compound. Our scientific results to date have landed us six licensing deals. All of this with less than 20 employees and $15M in spend. Other companies use similar resources to power a single disease program. We differentiate from other companies in this space in that our technology has been proven in over 18 disease areas.

We work with the best

Finally, you don’t have to look far to see who has associated themselves with twoXAR. Our pharmaceutical partners include heavyweights like Ono and Santen. Our management team previously held top level positions at Sony Biotechnology, Jazz Pharmaceuticals and Hewlett Packard. Our advisory board includes C-level execs from top pharma companies, as well as 30+ prestigious scientists from organizations such as Stanford University, Weil Cornell Medical and Dana Farber. Our company was the first startup to be accepted to both Stanford University’s StartX accelerator program and MIT’s prestigious STEX25, and we are backed by top tier investors including Andreessen Horowitz, Softbank, and OS Fund.

We are unique

These differentiators have contributed to twoXAR Pharmaceuticals success to date. twoXAR’s unique approach to drug discovery, utilizing our proprietary AI-driven platform and combining steps from the traditional drug discovery process, enables our company to accelerate the pace and success rate of discovering first-in-class compounds. With a pipeline of over 18 disease areas, twoXAR is working towards its mission to create new medicines for patients in need.



Aria Pharmaceuticals

Aria Pharmaceuticals is a preclinical-stage pharmaceutical company discovering and developing novel small molecule therapies for complex diseases.