Simultaneous analysis of millions of individual cells empowered with machine learning has potential to revolutionize treatments of many challenging human diseases including cancer, inflammatory and neurological disorders.

Biology has become a data science. For instance, recent single-cell profiling technologies are capable of measuring thousands of variables for each of hundreds of thousands to millions of cells in a single sample. Machine learning models are beginning to provide significant improvements in extracting information sensitively and with speed from such datasets which are often high-dimensional, sparse, and complex. Consequently, last year has seen enormous advances in deep learning applications in a variety of single-cell omics assays including genomics, transcriptomics, proteomics, metabolomics and multi-omics integration. It is highly timely to discuss the potential impact of insights generated by multi-omics machine learning platforms on the patient journey, clinical research and wider pharmaceutical sector through this mini-conference.

We will cover examples and potential applications of deep learning on single-cell and other high throughput datasets in the context of:

• Early diagnosis

• Differential diagnosis against overlapping etiologies

• Patient stratification for therapy

• Design of clinical trials

• Understanding disease mechanisms

• Drug discovery and repurposing

• Lifestyle decisions