Probabilistic logic learning from haplotype data

The analysis of haplotype data of human populations has received much attention recently. For instance, problems such as Haplotype Reconstruction are important intermediate steps in gene association studies, which seek to uncover the genetic basis of complex diseases. In this chapter, we explore the application of probabilistic logic learning techniques to haplotype data. More specifically, a new haplotype reconstrcution technique based on Logical Hidden Markov Models is presented and experimentally compared against other state-of-the-art haplotyping systems. Furthermore, we explore approaches for combining haplotype reconstructions from different sources, which can increase accuracy and robustness of reconstruction estimates. Finally, techniques for discovering the structure in haplotype data at the level of haplotypes and population are discussed.