This talk will focus on how to leverage AI to navigate large chemical spaces to support hit finding and lead optimization for challenge targets. It will describe how to use AI approaches to improve the standard library design, virtual screening, and hits expansion to support lead finding. It will also cover some concepts to leverage AI to rapidly generate testable hypothesis for lead optimization.

Target identification and validation is a critical step in the drug discovery pipeline. AI and machine learning (ML) techniques are increasingly being used for target discovery. This talk will provide an overview of AI/ML techniques in this space, including natural language processing, Bayesian networks, classical machine learning, and more recent deep learning methods such as graph neural networks and embeddings. We also discuss current research gaps and future opportunities.

We are entering a new phase of computationally-guided, informatics-intensive, collaborative drug discovery. We outline a digital chemistry strategy enabled by efficient exploration and scoring of vast, intelligently enumerated chemical space through the amplification of physics-based principles with machine learning. We describe the ‘Lab of the Future’ based on highly predictive models utilized across project teams. Case studies from drug discovery programs in preclinical and clinical stages.

As much as we hear about ML, not many are taking advantage of of it in their scientific work. By delivering ML-driven functionality for targeted use cases, modern software can turn the theoretical into practical tools. This presentation focuses on two use cases: protein structure prediction and strain parameter optimization. We’ll highlight how Benchling makes these tasks more efficient, and highlight the impact of customers currently taking advantage of ML as part of this everyday work.

Learn how to create impactful machine learning models and ensure early adoption into your daily workflow. We will share our experience working on a Blood-Brain-Barrier penetration machine-learning model using data from Reaxys.

Find out how to optimize model development through a cross-functional/organizational team approach and how Elsevier can supplement your efforts with data, support and services to enable an end-to-end development and deployment of machine-learning models.

Anti-drug antibodies developed against therapeutic proteins have been shown to reduce the medication efficacy, and in worst-case this immunogenicity can have safety implications for the patient. Understanding and predicting the immunogenic potential of therapeutic proteins has been a major challenge in the process of biotherapeutic drug discovery. In recent years, many AI methods have been applied to protein fragments to evaluate their immunogenicity potential and have shown encouraging accuracies. This presentation will provide an overview of popular AI techniques in this space, including natural language processing, position-specific scoring matrix, deep motif deconvolution, etc. Current use cases of such tools, research gaps, and future opportunities will also be discussed. 

Advanced digital microscopes have revolutionized biology & pharmaceutical research, generating massive volumes of data. But scientists can’t study what they can’t measure. The need to quantify biological characteristics or to calculate metrics has driven a need to seamlessly ingest, annotate, share, analyze, & archive datasets. Discovery Imaging Platform makes image datasets easily discoverable, accessible, transformable, analyzable, & ensures that derived numeric datasets are easily transferable to DL data analysis platforms.

Artificial Intelligence (AI) for drug discovery and development has entered a growth phase. This presentation will focus around the commercial aspects and call out major current and emerging application areas of AI/ML within the sector. AI-enabled competencies across big tech and start-ups will be covered followed by a brief assessment of the therapeutic area-wise activity and recent trends. Preview of the emerging ecosystem, partnerships, and selective case studies will also be discussed.

Cellarity uniquely combines high-resolution data, single cell technologies, and machine learning to encode biology, simulate interventions, and purposefully design breakthrough medicines. By focusing on the cellular changes that underlie disease instead of a single target, Cellarity’s approach uncovers new biology and treatments and is applicable to a vast array of disease areas.

The traditional process of drug discovery is complex, requiring the laborious experimental assessment of thousands of targets, hits, leads and candidates, making it expensive and time-consuming. Machine learning (ML) approaches provide methods that can improve the efficiency of the discovery process by formally integrating insights from data generated both in the public domain as well as internally. This talk will focus on case studies that demonstrate the application of ML to target validation and lead optimization and illustrate how machine learning methods can be used for decision making with quick informed predictions that can be rapidly validated by targeted experiments. 

In a large organization such as Sanofi, to better leverage predictive models developed by data scientists it is essential to have an operation model that allows us to move from “experiments” to impactful models leveraged by any. Sanofi is investing in developing key approaches that enable global operationalization of our predictive models with the goal of accelerating the discovery of large molecules.

The speed at which vaccines were launched in the past year has raised the bar, demanding accelerated innovation. The biopharma industry has leveraged innovative technologies ranging from the use of GPU-powered transformer models, federated learning platforms, and data fabrics, and has disrupted conventional models fueling co-innovation. AI is being used to design precision-engineered drugs and innovative strategies involving the use of computer vision are being leveraged to identify lead candidates for multiple diseases. Specific process analytical technologies (PAT) and intelligent supply chains are being developed for the manufacturing of highly complex and time and cost-intensive cell and gene therapies. While cloud-native platforms and high-performance computing are hyperscaling innovation, there are still significant concerns around data security and data ownership.