CORE SESSION STREAM 2: Research & Development
Data integration, the application of novel technologies, and machine learning methods are impacting all elements of translational research and the drug R&D process. With the rapid development of innovative data and technology approaches, challenges in adoption and integration need to be addressed at each stage.
Directors, Heads, VPs and scientists from pharma, biotech and research sectors in the following and related areas: computational and systems biologists, heads of R&D, innovation and specific development programs, translational leads, data analytics, bioinformaticians and software engineers.
AI, drug discovery & development, pharma, translational, multi-omic, in-silico modelling, machine learning.
TABLE 1: Applying Machine Learning to Drug Discovery and Repurposing
The time-consuming, often ill-fated and costly manner of drug development has encouraged the exploration of machine learning algorithms by drug development companies. The challenge of the high demand for exploring and analysing big data and the need for improvement at multiple steps in drug development process have the potential to be solved through the application of AI. Machine learning methods have also provided a new avenue in drug repurposing, shortening development timelines and costs significantly. The potential of machine learning needs to be fully realised, stripping away the hype, through collaboration of software developers and pharmaceutical leaders.
TABLE 2: Large-Scale Multi-Omics Data Integration and Analysis for Translational Research
Numerous high throughput -omics technologies have led to an explosion in the volume and richness of bio-data. We now need up-to date systems and infrastructure that allow for the integration, management and exploration of this data for translational research. Integration and analysis are critical steps in the understanding of disease and improvements in diagnosis and the development of better personalized therapies.
TABLE 3: Making Best Use of Statistical Modelling Methods of Tumours for Oncology Research
Complex mathematical tools and statistical modelling are already being used on large datasets from tumours, to find patterns of disease progression and tumour development in silico. However, these approaches can often be overwhelming for biologists and clinical scientists, who frequently lack a mathematical background. This hinders experimental possibilities and the benefits of in silico modelling for oncology research. Tools need to be developed that make computational and systems biology research more accessible for both researchers and clinicians, and to encourage use of in silico modelling for better understanding of tumour dynamics.
TABLE 4: The Role of Machine Learning in Identifying Novel Targets and Pathways
Advances in genomics have led to the creation of large data sets that can uncover the details of biological processes involved in human health and disease. Despite these advances, drug discovery and development remains a time-consuming, high-risk and incredibly difficult process, with a worryingly small success rate. Many of the answers to the challenges in this process remain locked inside the data. This efficiency and innovation gap in drug development needs to be addressed through innovative approaches, such as machine learning, to dramatically increase the success rate of lead generation for drug discovery, and to help build more robust drug discovery pipelines.