Challenges and Opportunities with Causal Discovery Algorithms: Application to Alzheimer’s Pathophysiology
Xinpeng Shen, Sisi Ma, Prashanthi Vemuri, György Simon, the Alzheimer’s Disease Neuroimaging Initiative, Michael W. Weiner, Paul Aisen, Ronald Petersen, Clifford R. Jack, Andrew J. Saykin, William J. Jagust, John Q. Trojanowki, Arthur W. Toga, Laurel Beckett, Robert C. Green, John C. Morris, Leslie M. Shaw, Zaven S. Khachaturian, Greg Sorensen, María C. Carrillo, Lew Kuller, Marc Raichle, Steven M. Paul, Peter J. Davies, Howard Fillit, Franz Hefti, David M. Holtzman, M. Marcel Mesulam, William C. Potter, Peter J. Snyder, Adam J. Schwartz, Tom Montine, Ronald G. Thomas, Michael Donohue, Sarah Walter, Devon Gessert, Tamie Sather, Gus Jiminez, Archana B. Balasubramanian, Jennifer Mason, Iris Sim, Danielle Harvey, Matt A. Bernstein, Nick C. Fox, Paul M. Thompson, Norbert Schuff, Charles DeCarli, Bret Borowski, Jeff Gunter, Matthew L. Senjem, David Jones, Kejal Kantarci, Chad Ward, Robert A. Koeppe, Norm Foster, Eric M. Reiman, Kewei Chen, Chet Mathis, Susan Landau, Nigel J. Cairns, Erin Franklin, Lisa Taylor‐Reinwald, Virginia Lee, Magdalena Korecka, Michal Figurski, Karen Crawford, Scott Neu, Tatiana Foroud, Steven Potkin, Kelley Faber, Sungeun Kim, Kwangsik Nho, Leon J. Thal, Neil Buckholtz, Marilyn Albert, Richard Frank, John Hsiao, Jeffrey Kaye, Joseph F. Quinn, Lisa C. Silbert, Betty Lind, Raina Carter, Sara Dolen, Lon S. Schneider, Sonia Pawluczyk, Mauricio Beccera, Liberty Teodoro, Bryan M. Spann, James B. Brewer, Helen Vanderswag, Adam Fleisher, Judith L. Heidebrink, Joanne Lord, Sara S. Mason, Colleen S. Albers, David S. Knopman, Kris Johnson, Rachelle S. Doody, Javier Villanueva‐Meyer, Valory Pavlik
Abstract
Causal Structure Discovery (CSD) is the problem of identifying causal relationships from large quantities of data through computational methods. With the limited ability of traditional association-based computational methods to discover causal relationships, CSD methodologies are gaining popularity. The goal of the study was to systematically examine whether (i) CSD methods can discover the known causal relationships from observational clinical data and (ii) to offer guidance to accurately discover known causal relationships. We used Alzheimer's disease (AD), a complex progressive disease, as a model because the well-established evidence provides a "gold-standard" causal graph for evaluation. We evaluated two CSD methods, Fast Causal Inference (FCI) and Fast Greedy Equivalence Search (FGES) in their ability to discover this structure from data collected by the Alzheimer's Disease Neuroimaging Initiative (ADNI). We used structural equation models (which is not designed for CSD) as control. We applied these methods under three scenarios defined by increasing amounts of background knowledge provided to the methods. The methods were evaluated by comparing the resulting causal relationships with the "gold standard" graph that was constructed from literature. Dedicated CSD methods managed to discover graphs that nearly coincided with the gold standard. For best results, CSD algorithms should be used with longitudinal data providing as much prior knowledge as possible.