3D density map

3D density map of 13,000 germ cells, as distributed in their gene expression PC1-PC2 space. Regions of higher cell density are shown as taller peaks. By Sue Hammoud, Chris Green, Qianyi Ma, Jun Li.

The unprecedented wealth of health data brings unprecedented opportunities and challenges to improve healthcare.  On the most microscopic level, we are now examining the genome one cell at a time, monitoring in real-time an individual’s health behavior, and prescribing medication based on the genetics and chemistry of each individual.  On the most macroscopic level, we can combine dozens of clinical record data for one patient, and can analyze the health data of an entire nation.  However, our ability to gain insight into such vast data lags significantly behind our ability to accumulate data.  The MIDAS Data-Intensive Health Science Research Hub aims to position UM researchers as national leaders as biomedical science embraces the opportunities and challenges brought about by advances in data science, catalyze the development of theories and innovative methodologies in data science, and their application to the entire spectrum of biomedical science.  The hub currently funds three projects.  With a variety of hub activities, MIDAS hopes to:

  • Disseminate tools and methods to empower campus-wide health research that integrates data science.
  • Build a collaborative network of Big Data health science researchers .
  • Form industry partnerships and transform research findings into clinical applications that have immediate impact on healthcare.

Michigan Center for Single-Cell Genomic Data Analytics

Single-cell genomics, rooted in single-cell sequencing, has great potential for providing insight into fundamental questions in biomedical science and drive new health science discoveries, such as: How many cell types and functional states are there in a given tissue?  What is the range of natural variation within a cell type and how is such variability affected by genetic and environmental factors?  What happens at the single-cell level during cell fate determination in the developmental process?  How does cellular heterogeneity with a tumor affect response to therapy and how can we address this with precision medicine?  The list is endless.  However, the explosive growth of single-cell sequencing technologies also brings new computational challenges.  One major challenge is the “sparse read counts data”: because of the minuscule amount of genetic material in a single cell, fragments of the genome are often missing from the sequencing read-out, and existing tools are ineffective in addressing this missing-data problem and piecing together reliable genomic information.

From Big Data to Vital Insights: Michigan Integrated Center for Health Analytics & Medical Prediction (MiCHAMP)

The recent explosion in health data has created unprecedented opportunities for healthcare improvement.  However, traditional analytics and existing computational platforms are poorly suited for handling the size and complexity of health data, leading to few real-world examples of ‘big data’ successfully impacting clinical care.   One core methodological challenge that currently limits health research is to analyze temporal patterns in longitudinal data for novel discovery and prediction. Although there exists an extraordinary volume of information on patients over time, temporal patterns are frequently overlooked in favor of simplistic, cross-sectional snapshots.  This project aims to develop methodologies for understanding longitudinal data, estimating time-varying parameters and predicting patient-specific trajectories.

Identifying Real-Time Data Predictors of Stress and Depression Using Mobile Technology

One robust finding in depression research is that life stress is the single most important trigger of depressive episodes. Understanding how life stress leads to depression has the potential to transform our ability to prevent and treat depression. Unfortunately, the capacity to capture the effects of stress accurately and in real-time has been limited because assessments of psychiatric phenotypes has traditionally relied on patients’ self-report of symptoms. Mobile electronic technology holds great promise in overcoming these limitations by capturing continuous, real-time, passive measures likely to be related to the progression from stress to depression. In this project, the research team will determine the dynamic relationships between mood, sleep, and circadian rhythms, and develop a micro-randomized intervention trial for depression under stress using an app platform that integrates mobile phones signals and wearable data. The team will build personalized models of sleep, circadian rhythms, physical activity, and mood, and assess and optimize the efficacy of objective mobile sensor to detect depression onset.  The researchers believe that when real-time data on an individual is fit to the right mathematical model, simple and insightful information can be easily communicated back to the end user as well as their caregiver and/or domain expert.