Mark Craven

Professor of Biostatistics & Medical Informatics

4775A Medical Sciences Center, craven@biostat.wisc.edu,  608.265.6181

Departments,

Biostatistics & Medical Informatics, Computer Science

Education

PhD (1996), University of Wisconsin,
Postdoctoral fellow at Carnegie Mellon University

Research Interests

We develop and apply machine-learning methods to infer models characterizing networks of interactions among genes, proteins, metabolites, clinical variables, environmental factors, and phenotypes of interest.

Lab Website

https://www.biostat.wisc.edu/~craven/

Representative Awards

  • 2001 NSF CAREER Award
  • 2011 Vilas Associates Award, University of Wisconsin-Madison
  • Director, CIBM Training Program
  • Director, BD2K-funded Center for Predictive Computational Phenotyping

Research

The focus of my research program is to develop and apply machine-learning methods to the problems of inferring models of, and reasoning about, networks of interactions among genes, proteins, metabolites, clinical variables, environmental factors, and phenotypes of interest. Current projects in my group are focused on (i) uncovering the intracellular networks involved in host-virus interactions, (ii) learning models of viral genotype-phenotype associations, (iii) automatically extracting molecular interactions and events from the biomedical literature, (iv) learning models to assess risk for clinical events such as asthma exacerbations and post-hospitalization VTEs using electronic health records and genetic data, and (v) modeling complex disease trajectories

Representative Publications  (Google Scholar | PUBMED)

  • Kiblawi, S. et al. Augmenting subnetwork inference with information extracted from the scientific literature. PLoS computational biology 15, e1006758, doi:10.1371/journal.pcbi.1006758 (2019).
  • Chasman, D. et al. Inferring host gene subnetworks involved in viral replication. PLoS computational biology 10, e1003626, doi:10.1371/journal.pcbi.1003626 (2014).
  • Hao, L. et al. Limited agreement of independent RNAi screens for virus-required host genes owes more to false-negative than false-positive factors. PLoS computational biology 9, e1003235, doi:10.1371/journal.pcbi.1003235 (2013).
  • Kawaler, E. et al. Learning to predict post-hospitalization VTE risk from EHR data. AMIA Annu Symp Proc 2012, 436-445 (2012).
  • Smith, A. A., Vollrath, A., Bradfield, C. A. & Craven, M. Similarity queries for temporal toxicogenomic expression profiles. PLoS computational biology 4, e1000116, doi:10.1371/journal.pcbi.1000116 (2008).
  • Noto, K. & Craven, M. Learning probabilistic models of cis-regulatory modules that represent logical and spatial aspects. Bioinformatics (Oxford, England) 23, e156-162, doi:10.1093/bioinformatics/btl319 (2007).