In today's world data is produced at a mind-boggling volume and pace. According to marketing reports 90 percent of the data in the world was produced within the past two years. With the recent advances in high-throughput technologies, OMICS data in the biomedical field would be one of the biggest contributor to this everlasting growth. These large volumes of data are widely accessible; however, they are mostly unstructured while having the potential to deepen our understanding of complex problems from disease outbreak to disease diagnosis and treatment. Statistical and machine learning methods can help expose the hidden value of unstructured data, improving prediction accuracy, and build predictive models that go beyond human performance. Recently, there is a surge of new techniques in the context of discovering latent structure in high dimensional data such as causal relations in multifaceted regulation networks, genotype and phenotype association, genomics signatures, and risk factors etc. However, it can still be a long way to obtain satisfactory results in scalable learning with increasing data size and complexity. Furthermore, a wide variety of technologies produce heterogeneous data shedding light on different aspects of complex biological systems. Emergence of a more complete picture of biological systems depends on successful methods for integration of data from these different perspectives.
With this workshop, we aim to encourage researchers to develop new methodologies, analytical models, and high-throughput computing workflow that best utilize various types of biomedical data in ways that meaningful structures present but hidden in data can be revealed.