Bootstrap Calibration of TRANSMIT for Missing Parental Genotype Data
Author Information
Author(s): Allen Andrew S, Collins Julianne S, Rathouz Paul J, Selander Craig L, Satten Glen A
Primary Institution: Duke University
Hypothesis
Can bootstrap calibration improve the accuracy of MAR procedures in the presence of informative missingness in parental genotype data?
Conclusion
The bootstrap-calibrated approach provides more reliable results than traditional MAR methods when dealing with informative missingness.
Supporting Evidence
- The bootstrap-calibrated and MAR-based inferences corresponded well on marker GATA25A04.
- Discrepancies were noted on markers GATA49C09 and ATC6A06, indicating the impact of informative missingness.
- Simulations suggest that even mild informative missingness can significantly affect MAR-based tests.
Takeaway
This study shows that when we don't have all the information about parents' genes, we can use a special method to make better guesses about their missing data.
Methodology
The study used a bootstrap calibration method to adjust MAR-based tests for informative missingness in parental genotype data.
Potential Biases
Potential biases may arise from the assumptions made regarding the relationship between missingness and genotype.
Limitations
The study's findings may not be generalizable beyond the specific population and markers studied.
Participant Demographics
Participants were nuclear families from the Framingham Heart Study, focusing on hypertensive individuals.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website