A practical guide to implementing AI
In this ebook, you’ll discover the key considerations which every leader needs to take in order to successfully implement AI in their drug discovery pipelines.
Cerella has the ability to work with and learn from all your data. To maximise the benefits of this, we’d advise ensuring that Cerella always has access to the most up-to-date and complete data. The best way to ensure this is to set up a live connection from Cerella to your data.
Customers who have taken advantage of this capability report improved performance, lower ongoing maintenance effort and better consistency of data.
Cerella supports the configuration of one or more data sources that can be used to provide data for the model-building process. While data from a flat file can be convenient during preliminary software evaluation, the true power of the software is only apparent with a live connection to the latest data. Cerella supports connections to commonly used databases such as Oracle and Postgres. It can also be customised to support a wide range of alternative data sources.
As a bonus, any data sources created for Cerella will also be visible to StarDrop users via the Query Interface tool.
We recognise that connection from a cloud-based application such as Cerella represents a challenge from a technical and data-security perspective. We have successfully managed similar connections for our customers using virtual private networks (VPNs). These ensure that data is never visible on the public internet. In addition, industry-standard encryption of all data is enforced in all our cloud-based software. Our DevOps team are experienced in the latest technology and will work with your IT team to ensure a safe and stable connection.
As with data connections to StarDrop, there is often a need to undertake some data preparation in order to produce input data that Cerella can work with most effectively. We can work with you to ensure that this is achieved in the most efficient way.
In some cases it is possible to organise the data in the database in such a way that Cerella can work with it directly. We are happy to advise you on how best to achieve that. However, in many cases, it is preferable to implement a customised version of Cerella that performs the initial data access steps and then prepares the data according to rules that you define. Our team would work closely with your scientists and informaticians to ensure the system generates the required data and associated metadata.
Optibrium’s focus has always been on developing new tools to extract the maximum knowledge from your data. We focus on cutting-edge science and state-of-the-art user interfaces to deliver this. However, we also recognise that these tools can only deliver the results you need if they are working on the best available data. We have devoted significant effort to developing multiple ways to achieve this in a streamlined way. We have successfully partnered with many customers and other organisations to deliver the right data in the right format, and look forward to discussing your particular requirements in future.
Chris earned a BSc and PhD in Chemistry from King’s College London, specialising in organic reaction mechanisms. After working at the Royal Society of Chemistry, he joined Inpharmatica in 2006, contributing to early StarDrop features like the Auto-Modeller. At Optibrium, he has driven key developments for StarDrop and, since 2018, has led custom projects integrating Optibrium software with clients’ in-house tools.
In this ebook, you’ll discover the key considerations which every leader needs to take in order to successfully implement AI in their drug discovery pipelines.
Is AI-guided drug discovery faster and cheaper? The evidence for this is, by definition, anecdotal. No one runs the same…
In this ebook we demonstrate our deployable AI discovery platform, Cerella™. Browse real-world stories of success from our collaborations with AstraZeneca, Genetech, Takeda Pharmaceuticals, Constellation Pharmaceuticals and many more.