Molecule AI Newsletter: May 2024
Discover the world of in silico drug discovery through Molecule AI's monthly newsletter. Join our journey of learning and impactful discoveries in bringing life-saving medications to patients faster.
Indian Pharma Attaining Global Capabilities with AI
Global capability centers, also known as GCCs, which are offshore units of multinational corporations that operate across the globe, are currently playing a crucial role in driving global business growth across industries. According to reports, India is home to more than 1,800 GCCs which employ over 1.3 million people. By 2030, the GCC market is estimated to exceed $100 billion, with 2500 GCCs across the country employing over 4.5 million people.
Cities like Bengaluru, Hyderabad, Delhi NCR, Mumbai, Pune, and Chennai are the most popular destinations offering a conducive environment for GCCs in India. However, tier-II towns such as Visakhapatnam, Jaipur, Vadodara, Kochi, and Chandigarh are becoming equally popular. Data reveals that US-headquartered firms account for the majority of the operational GCC footprint in the top 6 cities of India, followed by European firms (35%).
One of the key aspects of GCCs in India is the availability of a highly skilled workforce, as the country produces millions of graduates each year, trained across fields such as engineering, computer science, biotechnology and business management. Also, the cost of operating a capability centre in India is much lower than in developed countries, such as the US and the UK.
Sectors such as retail, automobile, and life sciences are viewing ‘Data and Analytics’ as a core function and thus using data-driven approaches to optimize decision-making. Data and Analytics-based GCCs are harnessing technologies like Artificial Intelligence (AI), Machine Learning (ML) and big data analytics, to analyse datasets for actionable insights, and for guiding business strategies, thereby transforming into innovation hubs.
Focusing on the pharma sector, the GCC market has grown considerably in the last few years. These centres, also known as global in-house centres (GICs), cater to various functions such as clinical trials, drug safety, regulatory filings, and drug discovery for global pharma giants. For instance, French pharmaceutical giant Sanofi has established its GCCs in Hyderabad to play a pivotal role in driving innovation, research, and development in healthcare, leveraging Sanofi’s extensive experience. Likewise, Swiss firm Novartis has a significant GCC presence in India with centres primarily located in Hyderabad and Bengaluru.
Another example is UK-based pharma company AstraZeneca which has over 3,100 employees across its innovation and technology centres in Chennai and Bengaluru which host 50% of the company’s technology operations. The focus of these centres is on Gen AI applications and immersive technologies to help doctors understand how the new drugs are working.
Very recently, Switzerland-headquartered Roche has opened its digital centre at Pune that can accommodate nearly 1,300 professionals focused on developing cutting-edge solutions using the latest technologies, including data and analytics, cloud computing, AI and ML.?
With AI adoption increasing within the pharma sector, companies are now tuning their respective GCCs to maximise the benefits of this technology. As a result, AI is predicted to disrupt and transform capability centres in the coming decades, and this calls for GCCs in India to position themselves as a model template for developing AI centres of excellence.
For instance, pharma GCCs can transform data into a strategic asset, making them more agile and competitive, with the help of AI. This technology can prove to be the ultimate problem-solving companion by creating a panacea through process automation, with AI-driven bots and algorithms adeptly handling repetitive and rule-based tasks. AI can minimize errors and accelerate operations, ushering in newfound efficiency, accuracy, and cost savings for the pharma GCCs.
In addition, GCCs can utilize AI’s pivotal role in enhancing customer support with AI-driven chatbots and virtual assistants responding instantly to customer inquiries, leading to elevated customer satisfaction and unwavering loyalty. Further, by leveraging AI for risk management and compliance, GCCs can fortify their operations and safeguard their invaluable reputation.
领英推荐
But this would require pharma GCCs to invest in developing a robust AI infrastructure to support the development and deployment of AI solutions at scale, responsibly and ethically. An equal focus would have to be on establishing a strong AI talent pool with a strong background of scientific knowledge of pharma R&D.
GCCs would also have to develop strong partnerships with startups, academia, and IT experts to keep abreast of the latest AI upgrades, and to foster a culture of knowledge sharing and collaboration. Looking ahead, we expect AI-based GCCs for the Indian pharma sector to drive the next wave of global growth.
Molecule GEN Highlights
In this issue of our newsletter, we present to you the MAI-Rings Filter (MAI-RF) module of Molecule GEN. This module offers ring recognition capabilities for hit-to-lead filtering, which is a core step in the drug discovery process. While the overall hit-to-lead filtering capabilities of the Molecule GEN platform include other standard filters based on medicinal chemistry properties and PASS/FAIL filters based on a large set of popular rules such as the Lipinski rule, the ring filter represents a cutting-edge filtering criterion which is as yet not implemented in other available in silico platforms, likely due to the challenges involved in developing accurate ring-recognition algorithms. Our implementation of this capability equips our users with a criterion which has been recognized in the literature as being highly predictive of the likelihood of a molecule being approved as a drug.??
MAI-RF is part of Molecule GEN, a web-based, modular drug-discovery platform being developed by MoleculeAI. Molecule GEN is envisioned as a go-to platform for the drug discovery community, due to its offering of diverse, user-friendly workflows such as the visualisation and analysis of target proteins, generation of high affinity hits against given targets, assessment of the med-chem and ADMET properties of molecules, lead optimisation, docking and molecular dynamics simulations.?
The MAI-Rings Filter (MAI-RF): Ring Recognition for Hit-To-Lead Filtering
The occurrence of ring structures in drug molecules has been studied by Shearer et al (Rings in Clinical Trials and Drugs: Present and Future | Journal of Medicinal Chemistry (acs.org)), and Taylor et al (Rings in Drugs | Journal of Medicinal Chemistry (acs.org)). They find that future drugs are very likely to contain rings from the set of ~380 ring structures already present in approved drugs, with only about six new ring structures being introduced every year. Moreover, the new rings added in any given year will likely be relatively minor modifications of ring structures existing until the previous year. Therefore, matching the rings in a potential new drug molecule against the set of rings in approved drugs becomes a useful early-stage predictor of the likelihood of success of the new drug candidate.?
MAI-RF offers a rapid and user-friendly means to perform this and other customisable matchings and comparisons. It is a cutting-edge novel tool that utilises a precise substructure-matching algorithm to search from a carefully curated database of rings derived from FDA-approved drugs. Accurate computational definition and matching of rings is not a trivial task, and substructure-matching algorithms provided by well-known platforms often cannot produce correct matches. For example, RDKit (RDKit) incorrectly identifies a match between the drug saxagliptin and the substructure piperidine, and ChemSpider (ChemSpider | Search and share chemistry) is unable to find the drug nedocromil as a match to the ring substructure within it (Figure 2). Similar deficiencies in substructure matching/searching are found in popular platforms such as PubChem (PubChem (nih.gov)) and ChEMBL (ChEMBL Database (ebi.ac.uk)), underscoring the necessity of designing a solution to this challenging problem.
With its robust substructure-matching algorithm, backed by a proprietary ring database which expands upon the work of Shearer et al., MAI-RF will prove to be an indispensable tool for the drug-discovery community. We provide MAI-RF as part of our broader suite of molecular filtration tools in Molecule GEN.
MoleculeAI in News