Highlights

-  600 AI BioTech companies and 100 corporations that develop drugs

- 1200 investors in the area of pharmaceutical and healthcare artificial intelligence

-- Database with Key Market players in the development of drugs using artificial intelligence

-- Comprehensive overview of investments in drug development companies

-- In-depth review of notable AI breakthroughs and pharma collaborations in 2021-2022

-- Overview of artificial intelligence methods that the famus companies use to developd drugs

Report at a Glance

This Artificial Intelligence for Drug Discovery Landscape Overview, Q2 2022” report marks the installment in a series of reports on the topic of the Artificial Intelligence (AI) application in pharmaceutical research industry that DPI have been producing since 2017.

The main aim of this series of reports is to provide a comprehensive overview of the industry landscape in what pertains adoption of AI in drug discovery, clinical research and other aspects of pharmaceutical R&D. This overview highlights trends and insights in a form of informative mind maps and infographics as well as benchmarks the performance of key players that form the space and relations within the industry. This is an overview analysis to help the reader understand what is happening in the industry nowadays and possibly give an idea of what is coming next.

Alongside investment and business trends, the report also provides technical insights into some of the latest achievements in the AI application and research.

Screenshot 2022-07-15 at 10.40.10.png

Pharma Efficiency: Challenges

10 years + $2.6 bln =  1 new drug

Screenshot 2022-07-15 at 11.33.06.png

It takes on average over 10 years to bring a new drug to market. As of 2014, according to Tufts Center for the Study of Drug Development (CSDD), the cost of developing a new prescription drug that gains market approval is approximately $2.6 billion. This is 145% increase, correcting for inflation, comparing to the same report made in 2003.

The solution to this problem comes from three key strategies: 

  • evolution of business models towards more collaboration and pipeline diversification early

  • implementation of AI as a universal shift towards data-centric drug discovery

  • discovery of new therapeutic modalities (biologics, therapies etc.)

0 - Effect on body

I - Safety in humans

II - Effectiveness at treating diseases

III - Larger scale safety and effectiveness

IV - Long term safety

Computer-aided Drug Design

Today's task for the pharma industry is to create a cheap and effective solution for drug development, companies apply various computational methods to reach that goal. Computer-aided drug design (CADD) is a modern computational technique used in the drug discovery process to identify and develop a potential lead. CADD includes computational chemistry, molecular modeling, molecular design and rational drug design

Screenshot 2022-07-15 at 11.42.58.png
Screenshot 2022-07-15 at 13.25.47.png

Modern computational structure-based drug design has established novel platforms that mostly have a similar structure for testing drug candidates. The usage of AI can simplify and facilitate the drug design from filtering datasets for appropriate compounds to advanced lead modification and in silico testings.

Application of AI for Advanced R&D to Address Pharma Efficiency Challenges

Screenshot 2022-07-15 at 13.41.09.png

Accelerated development of new drugs and targets identification

  • Identify novel drug candidates

  • Analyze data from patient samples 

  • Predict pharmacological properties 

  • Simplify protein design

Targeted towards personalized approach
and optimal data handling

  • Optimize clinical trial study design

  • Patient-representative computer models 

  • Define best personalized treatment 

  • Analyze medical records 

  • Improve pathology analysis

Time- and resources-efficient information management

  • Generate insights from thousands 
    of unrelated data sources

  • Improve decision-making

  • Eliminate blind spots in research

Searching for new applications of existing drugs at a high scale

  • Rapidly identify new indications

  • Match existing drugs with rare diseases

  • Testing 1000+ of compounds in 100+ of cellular disease models in parallel

Optimization of experiments and data processing

  • Reduce time and cost of planning

  • Decode open- and closed-access data 

  • Automate selection, manipulation, and analysis of cells

  • Automate sample analysis with a robotic cloud laboratory

Pharma's “AlphaGo Moment”

Screenshot 2022-07-15 at 16.52.37.png

Notable Breakthroughs in AI for Pharma

Screen Shot 2022-07-17 at 12.58.50 PM.png

Technological Advancements Defining the Market

Insilico Medicine achieved an industry-first fully AI-based Preclinical Candidate. The initial hypothesis was built via DNN analysis of omics and clinical datasets of patients. After that company used its AI PandaOmics engine for target discovery, analyzing all relevant data, including patents and research publications with NLP algorithms. In the next step, Insilico has applied its generative chemistry module (Chemistry42) in order to design a library of small molecules that bind to the novel intracellular target revealed by PandaOmics. The series of novel small molecules generated by Chemistry42 showed promising target inhibition. One particular hit ISM001 demonstrated activity with nanomolar (nM) IC50 values.

When optimizing ISM001, Insilico managed to achieve increased solubility, good ADME properties, and no sign of CYP inhibition — with retained nanomolar potency. Interestingly, the optimized compounds also showed nanomolar potency against nine other targets related to fibrosis. The efficacy and good safety of the molecule led to its nomination as a pre-clinical drug candidate in December 2020 for IND-enabling studies. The phase I clinical trial for the novel drug candidate is planned for December 2021.

Screen Shot 2022-07-17 at 1.08.11 PM.png

When optimizing ISM001, Insilico managed to achieve increased solubility, good ADME properties, and no sign of CYP inhibition — with retained nanomolar potency. Interestingly, the optimized compounds also showed nanomolar potency against nine other targets related to fibrosis. The efficacy and good safety of the molecule led to its nomination as a pre-clinical drug candidate in December 2020 for IND-enabling studies. 

MindMap

1.png

Comparison of Top-40 Leading AI for Drug Discovery Companies Expertise in Drug Discovery R&D

Screen Shot 2022-07-17 at 12.51.03 PM.png

To learn more about leading companies check our dashboard

525 Artificial Intelligence companies: Regional Distribution

1130 Investors: Distribution per Country

50 Leading Investors: Distribution per Country

Big Pharmas’ AI-focused partnerships till Q2 2022

In this report we have profiled 525 actively developing AI-driven biotech companies. A steady growth in the AI for Drug Discovery sector can be observed in terms of substantially increased amount of investment capital pouring into the AI-driven biotech companies ($2.28B in HY 2020 against $2.93B in HY 2021), the increasing number of research partnerships between leading pharma organizations and AI-biotechs, and AI-technology vendors, a continuing pipeline of industry developments, research breakthroughs, and proof of concept studies, as well as exploding attention of leading media and consulting companies to the topic of AI in Pharma and healthcare. 

Some of the leading pharma executives increasingly see AI as not only a tool for lead identification but also a more general tool to boost biology research and identify new biological targets and develop novel disease models. 

The main focus of AI research for today is still on small molecules as a therapeutic modality. 

Screen Shot 2022-07-17 at 6.41_edited.jpg

To learn more about collaborations check our dashboard

Business Activity

The business activity has been increasing in the pharmaceutical AI space over Q1 2021 - Q2 2022, judging by an increased number of transactions and partnership announcements in this period.

The most significant deals and collaborations in include:

Screen Shot 2022-07-20 at 4.07_edited.jpg

To learn more about business activity check our dashboard

Dynamics of Investments in AI in Pharma

There has been a substantial increase in the amount of capital invested in AI-driven pharma companies since 2014. During the last eigth years, the annual amount of investments in 600 companies has increased by almost 16 times (to $115.84B in total as of June 2022). In 2021, the flow of investments increased by 40% compared to the previous year which identifies strong investors’ (foremost VCs) interest in this field regardless of risks.

To learn more about the investments in AI in DD check our dashboard

Top 10 AI in Pharma Companies by Total Investments in 2021

The chart shows the top 10 AI-driven drug discovery companies sorted by the total funding raised by the end of  June 2022. The amounts of investments are shown in $.

XtalPi, a leader in artificial intelligence and precision medicine, is now at the top of the list abd has the total funding raised to $791M. Insitro, a drug discovery and development startup that utilizes machine learning and biology to transform drug discovery, could finance $743M in capital market. ThoughtSpot, Tempus and Neuromora are new companies due to late-stage mega-rounds during the 2021.

AI in Pharma IPO in 2021

In 2021 new public companies have successfully closed their IPOs. As for now, almost all these companies are showing a slight decline, which is typical for new pharmaceutical companies, especially with the negative net income. All IPOs took place in the USA. All companies have beta smaller than 1 (although positive), which means that AI in pharma stock prices move following the general market, yet the degree of such “movements” is lower. Major adverse market events in 2020-2022 did not significantly affect AI in pharma sector. The industry’s features remain to play a designative role in the overall market volatility.

Screen Shot 2022-07-20 at 4.26.27 PM.png

To learn more about publicly traded companies check our dashboard

Companies related to AI-Pharma

    AI in the pharma sector is an integral part of the contemporary pharmaceutical industry. AI-Pharma sector, defined broadly, is not limited to AI companies but includes also pharma, tech, chemistry corporations, and CROs that are engaged in collaborations with AI startups, including but not limited to Mergers & Acquisitions, scientific research, partnerships, and so on.Hence the companies chosen are better described as AI-related or AI-aiming than AI-based solely.

    The number of new partnerships between pharma companies and AI companies is ever increasing across the whole industry. On the one hand, AI-focused companies may spend a few years developing all software and tools which pharma companies do not have. On the other hand, large companies, mainly public ones, have a solid understanding of their science, and extensive experience in the industry and regulatory field, and they are ready to share the risk.

    In this chapter, we introduce the list of top corporations related to AI-Pharma that were selected based on the analysis of their R&D, financials, and collaborations with the most promising and advanced AI-Pharma startups.

Screen Shot 2022-07-20 at 4.37.58 PM.png

To learn more about publicly traded companies related to AI-Pharma check our dashboard

Notable R&D Use Cases of AI Application in Biopharma

How Standigm Accelerates Drug Discovery using AI?

1-removebg-preview (2).png

Standigm is a workflow AI-driven drug discovery company headquartered in Seoul, South Korea and subsidiarized in Cambridge, UK. Standigm has proprietary AI platforms encompassing novel target identification to compound design, to generate commercially valuable drug pipelines. The company has established an early-stage drug discovery workflow AI to generate First-in-Class lead compounds within seven months. o date, Standigm is running 42 in-house or collaborative pipelines for drug discovery using the workflow AI technology. One of the company's pipelines is expected to enter a pre-clinical stage in 4Q 2021. me. It's easy.

Standigm BEST is a novel compound generation platform, which can investigate lead compounds whenever target or ligand information is lacking or enough.

Standigm ASK is a customizable, AI-aided drug target identification platform, prioritizing disease-target relationships and providing evidence-based results through an interactive user interface.

1.png
1.png
1.png

How Strados Labs Uses AI in R&D?

1.png

Strados Labs enters the Pharma and Life Science market with a Respiratory Management Solution that includes the only FDA-cleared, RESP biosensor which acquires lung sound acoustics wireless and hands-free, making it a perfect fit for clinical research to measure patient response to new drugs by objectively collecting coughs and other lung sounds discreetly, comfortably, and securely in a streamlined way, while having access to data for post-processing and analysis.

Screenshot 2022-07-21 at 11.19.30.png

Strados Labs — a respiratory management solution, which brings innovation at the intersection of lung biomarkers, patient centricity, and machine learning. The industry of life sciences can largely benefit from the enhancement of pulmonary care monitoring capabilities provided by Strados Labs to gain insight into patient drug response by analysis of longitudinal lung acoustics.

Screenshot 2022-07-21 at 15.29.04.png
1.png

The Strados Respiratory Management Solution is the world's first FDA-cleared lung sound platform with a proprietary wireless biosensor, RESP, that is passive, patient-friendly, and clinically validated to acquire lung sounds in the real world.

Today Strados Labs has a unique opportunity to stand as a leader in Respiratory Health: their clinically validated bioacoustic library of sounds and AI engine is the world’s largest entirely hands-free, clinical-grade dataset enabling Strados Labs to be the standard bearer of acoustic digital biomarkers for clinical research and respiratory care globally.

Screenshot 2022-07-21 at 15.43.57.png

For instance, Strados Labs RESP fits perfectly into decentralized trials allowing remote patient access by unlocking lung sound data and putting it into the hands of the entire research team via the cloud.  Making decentralized respiratory trials scalable and able to develop entirely new insights about respiratory status without episodic patient interaction.

Screenshot 2022-07-21 at 15.47.32.png

Strados Cloud: company’s passive and longitudinal bioacoustics insights allow them to build a more complete picture of the subject’s respiratory status leading to better trial outcomes.

Screenshot 2022-07-21 at 15.50.46.png

How Antiverse Engineers the Future of Drug Discovery?

Antiverse is a new type of antibody discovery company accelerating drug development. The Antiverse platform exists at the intersection of structural biology, machine learning and medicine to enable breakthroughs to happen more quickly and cost-effectively.

Screenshot 2022-07-22 at 16.50.00.png

Antiverse prevents diversity loss during amplification to uncover more diverse and rare antibodies.

Antiverse provides more candidates by analysing NGS data, clustering on multi-dimensional space, and selecting based on sequential and structural grouping. 

1.png

Antiverse is recognized as one of the top biotech startups in the UK with our antibody discovery service already in use by big pharma. The main feature of the company is 10x Diversity with AI-Augmented Drug Discovery.

Existing antibody discovery methods are well-developed and often effective at discovering binders. But when there is a need to find the best possible candidate, or when finding a suitable candidate is hard with current methods, the options are limited and often costly.

 

Antiverse uses next-generation sequencing (NGS) to extract more data from existing workloads. The AI-Augmented Drug Discovery platform and trained models analyse the statistics gained from thousands of experiments. These outputs are compared against known data in order to select best candidates. 

Screenshot 2022-07-22 at 17.13.12.png

The Antiverse AI-ADD system found each and every cluster identified by other methods, plus more. These additional clusters contained rare and unique sequences.

Screenshot 2022-07-22 at 17.16.02.png

How Arctoris Accelerates Drug Discovery using AI?

Drug discovery is undergoing massive and rapid change - the rise of Artificial Intelligence and Machine Learning for Drug Discovery and the evolution of robotics-centric companies in the biomedical research space has enabled a new generation of companies to emerge: data-powered drug discovery companies that combine automation and data science. 

1.png

Arctoris is one of them: a biotech platform company with operations in Oxford, Boston, and Singapore, leveraging its fully automated plat- form for drug discovery.

The company was founded by an oncologist and a medicinal/ synthetic chemist, with the goal to accelerate the discovery and development of new therapies by harnessing the power of technology and combining it with deep industry expertise.

 

The core thesis of the company is that better data leads to better decisions, and that in order for drug discovery programs to develop and meet the next milestone faster and with higher chance of success, the underlying data must be rich, reliable, and reproducible. According to Arctoris, the status quo is no longer enough: in order to develop the best drugs, industry leaders have to rethink how they can improve their decision-making, powered by better data. 

 

Having developed a suite of proprietary technologies across robotics and data science/ AI/ ML, Arctoris is a leader in this new and rapidly evolving field. 

1.png

The greatest challenge in AI-driven and ML-powered drug discovery is access to well structured, fully annotated, reproducible and robust data. Arctoris leverages the power of robotics to generate vast amounts of ML-ready data that enable better decisions - thereby significantly accelerating timelines from target to hit, lead, and candidate.

1.png

Both quality and speed are achieved by combining precision robotics with a unique data science platform and world-class drug discovery expertise from biotech and pharma veterans. 

 

Arctoris tracks all experimental outputs in full depth, including the capture and analysis of extensive metadata – temperature, humidity, CO2, reagent provenance and batch ID among many others. At the same time, the platform enables automated QA/ QC processing, applying statistical tools to ensure full reliability and validity of all results. 

 

Thereby, Arctoris ensures superior data to be generated in accelerated timeframes, leading to better decisions taken earlier - in human-powered but especially in AI/ ML-driven programs, thanks to training of AI models with the best possible data. 


Taken together, Arctoris has developed a unique technology platform based on robotics and data science that powers drug discovery programs both in the company’s internal pipeline and in partnerships with biotech and pharma companies worldwide.

1.png
  • Analysis of target expression and target half-live by quantifying protein turnover and route to degradation

  • Investigation of target function (changes in phenotype, pathways, gene expression, etc.) via cell-based and molecular biology readouts

  • Advanced insights into effects of target modulation by employing complex model systems such as organoids, primary cells, etc.

  • Machine-learning guided screening set selection and hit evolution

  • In silico and in vitro screening and profiling

  • Biophysical screening/ profiling and FBDD

  • Rapid synthetic hit expansion and diversification incl. use of CADD

  • Kinetic and mechanistic biochemistry/ enzymology and biophysical quantitation of target engagement energetics & kinetics 

  • Protein science and (co)crystallography for SBDD

  • Pharmacokinetics and pharmaco- dynamics (PK/ PD) & safety pharma- cology

  • In-depth pharmacokinetics, including ADME, drug-drug interactions, metabolite profiling, concentration time profiles

  • Comprehensive acute toxicology assess- ment, incl. single dose and repeated dose to determine MTD and NOAEL

  • Additional toxicology studies (e.g. repro- ductive and developmental toxicity, etc.)

  • Pharmacokinetics and pharmaco- dynamics (PK/ PD) & safety pharma- cology

  • In-depth pharmacokinetics, including ADME, drug-drug interactions, metabolite profiling, concentration time profiles

  • Comprehensive acute toxicology assess- ment, incl. single dose and repeated dose to determine MTD and NOAEL

  • Additional toxicology studies (e.g. repro- ductive and developmental toxicity, etc.)

How Genomenon Uses AI in R&D?

1.png

Genomenon is an AI-driven genomics company that organizes the world’s genomic knowledge to accelerate the diagnosis and development of treatments for genetic disease.

Genomenon’s Prodigy™️ Genomic Landscapes deliver a profound understanding of the genetic drivers and clinical attributes of any genetic disease and support the entire drug development process, from discovery to commercialization. 

Genomenon’s main focus therapeutic areas are rare diseases, genetic diseases, and hereditary and somatic cancers.

1.png

Genomenon’s Prodigy™️ Genomic Landscapes use a unique combination of proprietary Genomic Language Processing (GLP) and expert, scientific review to provide an evidence-based foundation for all stages of the drug development process. These landscapes can be completed at the disease, gene, variant, or patient level, and are maximally comprehensive as a result of GLP. Genomic Landscapes are also rapidly produced using an AI-assisted curation engine that expedites manual review of the data indexed by GLP.

Genomic Language Processing (GLP) is a novel technology that systematically extracts and standardizes genomic and clinical information from the medical and scientific literature. Designed specifically to recognize this complex genomic information, GLP provides superior sensitivity compared to traditional methods, finding more variants and subsequently, more patients. Genomenon’s database, built using GLP, currently contains over 14.8 million variants, 8.8 million full-text articles, and 3 million supplemental datasets.

1.png

In collaboration with Alexion, AstraZeneca’s Rare Disease group, Genomenon applied its AI technology to help accelerate the genetic diagnosis for rare disease patients. Genomenon’s novel combination of AI-powered Genomic Language Processing and expert review identified significantly more pathogenic variants associated with Wilson disease.

 

Genomenon’s AI-driven approach identified 3.7x more evidence-supported, pathogenic/likely pathogenic variants for ATP7B – a gene associated with Wilson disease – compared to the crowd-sourced database, ClinVar.  This significantly expands the resources available to healthcare providers to make more informed diagnostic decisions.

 

1.png

Genomenon’s AI-driven approach identified 3.7x more evidence-supported, pathogenic/likely pathogenic variants for ATP7B than ClinVar.

We predict that this will improve the diagnosis of people living with Wilson disease by improving the ability to interpret genetic testing results.

How GATC Health Uses AI in R&D?

1.jpg

GATC Health is an AI company that accelerates the drug discovery and development process. The company proposes highly efficient services for pharma companies reducing the risk in the drug discovery process. GATC Health develops an end-to-end drug development cutting-edge AI-based platform. The platform assists in earlier disease detection, identify the disease biology, create new drug and therapeutic solutions, simulate in-silico clinical trials and feedback loop for in-vitro and in-vivo testing. 

 

GATC’s Platform combines massive volumes of disease-specific data and proprietary AI solutions to replicate human biology in discovering and validating novel drugs. The company develops a revolutionary approach to drug discovery, drastically improving efficiency and time for clinical development.

Screenshot 2022-07-26 at 13.55.18.png
  • Identifies the causal relationship between the biomarkers and the disease to illuminate insights into the disease.

  • AI-assisted compound discovery is used to produce a set of novel treatment compounds.

  • The targets and compounds are prioritized and documented for pre-clinical testing.

Diagnostic Biomarker Discovery
  • Diagnostic biomarkers are discovered on a dataset.

  • Biomarkers mathematically assessed for causal and effect impacts.

  • Validated causal biomarkers and pathways are simulated and evaluated by AI-assisted database models and human expertise.

  • A final set of treatment targets emerges.

Drug Compound Discovery
  • Pre-clinical de-risking of the drugs through through the development of new therapeutics in-silico and in-vivo clinical studies with more comprehensive analysis.

  • Ensuring higher levels of success as the drug progresses through FDA Trials.

  • Eliminate the majority of the risk and cost associated with resolving disease.

Pre-Clinical De-Risking of Drug
Screenshot 2022-07-26 at 14.08.57.png

To learn more about technological insights these and other companies use check our dashboard

Selected Pharma AI Industry Developments Q1 2020 — Q2 2022

Screenshot 2022-07-26 at 14.15.03.png

To learn more about Pharma AI Industry development check our dashboard

Screenshot 2022-07-08 at 16.45_edited.jpg

Artificial Intelligence for 

Drug Discovery

Landscape Overview Q2 2022

This report aims to provide a comprehensive overview of the industry landscape in what pertains adoption of AI in drug discovery, clinical research and other aspects of pharmaceutical R&D. This overview highlights trends and insights in a form of informative mind maps and infographics as well as benchmarks the performance of key players that form the space and relations within the industry.