Artificial intelligence in the world of drug discovery

Page 1

11/1/2019

Artificial intelligence in the world of drug discovery

Arti cial intelligence in the world of drug discovery Usm systems Nov 1 ¡ 6 min read

Research and development (R&D) is a long and expensive process. However, companies have recently begun to understand the importance and benefits of using AI services in their work.

https://dp9bxf2pat5uz.cloudfront.net/wp-content/uploads/shutterstock_1040985685-e1562161895771.jpg

Walker-Osborne advises companies from a legal perspective on the use of AI in their drug discovery processes and represents AI suppliers in the pharma sector. Over the last 30 years, there has been a bit of a machine learning element called AI now (often narrow https://medium.com/@fugenxmobileapp/artificial-intelligence-in-the-world-of-drug-discovery-a49aedad3d18

1/5


11/1/2019

Artificial intelligence in the world of drug discovery

AI is designed and trained for a specific task), but more sophisticated (sometimes referred to as “strong AI”) is now commonly implemented. “The use of AI in drug discovery is” one of the most obvious applications of AI, ”she says. In other applications, having a fast and accurate analytics platform allows companies to identify drug models and processes for drug discovery. A recent announcement that the top 10 companies are using AI to share their data from AI innovation as part of the Machine Learning Ledger Orchestration for Drug Discovery (MELODY Project) indicates that it will be increasingly used in the future. This project is the first of its kind and allows shared organizations to learn from shared data and improve AI through the Okin program. Walker-Osborne explained that AI can be used in silico design, development data development, big data analytics, study risk assessment, patient matching, and more to identify the target. For these reasons, it is being trialed and/or implemented by many companies, and some are developing their own advanced data analytics platforms that use AI. She said the above areas may be obvious, but “AI can also be used to summarize in plain English and scientific documents and help provide a starting point for companies’ own papers.” Yet there is a human element, disc discovery, capturing key points and points from data, and finding patterns AI can be very useful Continues. The latest collaboration between Eksencia and Celgene is a collaborative effort for drug discovery in oncology and autoimmunity. They agreed to partner for three years to accelerate the discovery of small therapeutic drug candidates. Key trends Most companies working on drug discovery are trying to use AI, and Walker-Osborne does this to some advantage, including the ability to work with AI compared to humans. Walker-Osborne commented that for the region, “speed to market is a game-changer from a cost and time perspective.” Eli Lilly’s new partnership with AtomWise helps identify novel target proteins to support drug discovery efforts. The goal is to speed up target identification using Eli Lilly's information.

https://medium.com/@fugenxmobileapp/artificial-intelligence-in-the-world-of-drug-discovery-a49aedad3d18

2/5


11/1/2019

Artificial intelligence in the world of drug discovery

Data sharing is another emerging trend. “The whole premise of AI is that you need high quality and good datasets for the machine to learn and do anything Another trend I see is that people are still very willing to share their [information] on the machine. To learn for the better, ”Walker-Osborne said. The Machine Learning for Pharmaceutical Discovery and Synthesis Consortium at the Massachusetts Institute of Technology (MIT) is a data-sharing program that includes companies such as GlaxoSmithKline (GSK), AstraZeneca and Eli Lilly. Their progress includes automating the design of molecules to accelerate drug development. With a lot of interest in AI, it is growing fast; New improvements have resulted in finetuning towards specific goals. Walker-Osborne argues that AI “can do very intelligent and fast things with datasets, but it can be more successful when it is focused on imaging for oncology, accelerating molecular discoveries or detecting compounds.” By using AI in a particular area, drug discovery can be very effective. She explains that it is the current trend and “absolutely critical” in accelerating drug discovery processes that often focus on specific applications of AI with leading pharma companies. Protecting intellectual property and data There are many legal issues with adopting AI. Intellectual property (IP) is the most critical around. Walker-Osborne explains: AI technologies evolve through use. Repeatedly training the AI system using a new data set will lead to a new model that will modify features and behavior. This creates tension in the traditional analysis of front and backgrounds IP, which is commonly applied to R&D collaborations. The trained model is clearly an improvement over the background IP and AI partner’s background IP. Therefore, whether or not the AI partner has the intellectual property that arises and any derivative data arises, whether or not the trained model/derivative data can be used outside the collaboration (and with no responsibility for the pharma partner). There is no one rule here. This is ultimately a point of agreement between the parties. “ She continues: “Where there is further research effort to derive output data from the AI platform, there will be additional creative and technical input that generates IP. In such cases, we recommend that the ownership and exploitation rights of the wider project https://medium.com/@fugenxmobileapp/artificial-intelligence-in-the-world-of-drug-discovery-a49aedad3d18

3/5


11/1/2019

Artificial intelligence in the world of drug discovery

product be treated in the same way as other R&D collaborations, and should be dealt with separately. “ Data privacy and cyber-security In the UK, the Office of the Information Commissioner (ICO) is issuing various useful guidelines around the use and auditing of AI, where personal data considerations are concerned. The guidance that emerges from the ICO in this area and redirects the laws around the world is almost the same. Says Walker-Osborne: “As with any R&D activity involving data, the rights/consents need to be considered for confidential/proprietary information and data privacy issues, and the roles and responsibilities of the parties are clearly defined. In many ways, the analysis is no different from the more traditional pharma R&D, and pharma companies take the security of their data very seriously. It is important for AI suppliers to provide security. Depending on the contribution, the nickname data sets may be used. However, there are data privacy issues to be solved (including the need to ensure that a person cannot be traced back). “ Cybersecurity is vital for AI to keep data secure. Walker-Osborne explained that cybersecurity is “an ongoing difficulty for all businesses, but I’m glad that all parties are taking this issue together to protect data.” Challenges of AI Introducing AI into Pharma and Drug Discovery Procedures In addition to the above, there are many legal challenges (usually overcome). As Walker-Osborne explains, “The main challenge is probably commercial — getting the right and big data sets, adopting the right AI, training the right way, making sure no bias is added or expanded in the process.” One of the biggest legal challenges with the application of laws to the adoption of AI, Walker-Osborne explained, “is that they are often very country-centric. These pharma companies are very global companies, so they need to think carefully about the laws that apply to their current situation, including where the AI platform is, where it is used, where the data is coming from, and where. Innovations dissolve and apply all of those laws. “ https://medium.com/@fugenxmobileapp/artificial-intelligence-in-the-world-of-drug-discovery-a49aedad3d18

4/5


11/1/2019

Artificial intelligence in the world of drug discovery

Another legal challenge is the IP mentioned above. For example, in the US, AI treatment in IP is very different from the UK and Europe and requires more clarity in places. ”However, the rate at which AI has evolved and the potential of the UK is not fully addressed by the UK government’s position. Walker-Osborne explained, “This year, we have a policy, monetary investment and change We have seen the emergence of transparency, but in my view, there is still much to be done, especially in the area of monetary investment… We need to find a balance between the freedom of use of AI in a positive environment and curb many restrictions and innovations. ” AI can be used in so many fields, and I think there is going to be a shortage of real AI expertise Skills possessed by industry workers pose another challenge in introducing AI. Workers may need to re-train or learn new skills to use complex machines. Walker-Osborne commented, “A lot of the AI community say there’s no problem with the re-skilling challenge… but I’m not sure. AI can be used in so many fields that I think there is going to be a shortage of true AI expertise. However, “there will be a significant increase in new roles — ethics officers, data scientists, etc.” “ So what does the future look like? Walker-Osborne claims that AI will become the core of pharma and is a key step in the R&D and drug discovery process. Walker-Osborn believes that for drug discovery, there will be a tipping point, where “adopting AI will get more affordable as with any technology… it will become very much part of the process and — with the right guide rails around its use — I believe that is a good thing.”

Machine Learning

Arti cial Intelligence

Ai Solution

Ai Services

Ai In Drug Discovery

About

https://medium.com/@fugenxmobileapp/artificial-intelligence-in-the-world-of-drug-discovery-a49aedad3d18

Help

Legal

5/5


Turn static files into dynamic content formats.

Create a flipbook
Issuu converts static files into: digital portfolios, online yearbooks, online catalogs, digital photo albums and more. Sign up and create your flipbook.