Impact of Missing Data on Genetic Testing Validity
Author Information
Author(s): Guo Chao-Yu, Heather J Cordell, Mariza de Andrade, Marie-Claude Babron, Christopher W Bartlett, Joseph Beyene, Heike Bickeböller, Robert Culverhouse, Adrienne Cupples, E Warwick Daw, Josée Dupuis, Catherine T Falk, Saurabh Ghosh, Katrina A Goddard, Ellen L Goode, Elizabeth R Hauser, Lisa J Martin, Maria Martinez, Kari E North, Nancy L Saccone, Silke Schmidt, William Tapper, Duncan Thomas, David Tritchler, Veronica J Vieland, Ellen M Wijsman, Marsha A Wilcox, John S Witte, Qiong Yang, Andreas Ziegler, Laura Almasy, Jean W MacCluer
Primary Institution: Boston University and National Heart, Lung and Blood Institute
Hypothesis
How does informative missingness affect the validity of the transmission/disequilibrium test?
Conclusion
The study found that the transmission/disequilibrium test is not valid when offspring genotypes are missing informatively.
Supporting Evidence
- The study derived the conditional distribution of ascertained triads allowing informative missingness.
- Simulations showed inflated type I error and/or reduced power for the TDT when offspring genotypes were missing informatively.
- The TDT and HRR were valid tests under certain missing patterns but invalid under others.
- Excluding incomplete triads from analysis led to invalid results for the TDT and HRR.
Takeaway
If we don't have all the information about parents and kids in a genetic study, we might get the wrong answers about diseases.
Methodology
The study derived the conditional distribution of ascertained triads allowing informative missingness and evaluated several tests under these scenarios using simulations.
Potential Biases
The assumption of missing data being completely at random may not hold true, leading to potential biases.
Limitations
The study primarily focused on specific missing patterns and may not generalize to all genetic studies.
Participant Demographics
Unrelated nuclear families with affected siblings were used in the simulations.
Want to read the original?
Access the complete publication on the publisher's website