Segregation analysis with uncertain ascertainment: application to Fanconi anemia

Area of Science:

• Genetics
• Biostatistics
• Computational Biology

Background:

• Accurate genetic parameter estimation is crucial for understanding inheritance patterns.
• Ascertainment bias can complicate genetic analyses, especially when data on ascertainment is unavailable.
• Bayesian inference offers a flexible framework for handling complex statistical problems in genetics.

Purpose of the Study:

• To develop a Bayesian statistical solution for genetic segregation analysis without ascertainment information.
• To infer segregation probability and the probability of sporadic cases by marginalizing the ascertainment probability.
• To validate the method using simulated and real-world genetic data.

Main Methods:

• A Bayesian approach was employed to model genetic segregation.
• Posterior marginal distributions were used to integrate out the nuisance parameter (ascertainment probability).
• The method was tested on both simulated datasets and original Fanconi anemia data.

Main Results:

• The developed Bayesian method performed well on both simulated and real data.
• Inferences regarding segregation probability and sporadic case probability were successfully obtained.
• Analysis of Fanconi anemia data confirmed a monogenic autosomal recessive inheritance pattern.

Conclusions:

• The proposed Bayesian solution effectively addresses the challenge of unknown ascertainment in segregation analysis.
• This method provides reliable estimates of genetic parameters, facilitating accurate mode of inheritance determination.
• The findings support a monogenic autosomal recessive inheritance for Fanconi anemia, advancing our understanding of the disease.