Kari Vasko

Learn More
We introduce a new method for linkage disequilibrium mapping: haplotype pattern mining (HPM). The method, inspired by data mining methods, is based on discovery of recurrent patterns. We define a class of useful haplotype patterns in genetic case-control data and use the algorithm for finding disease-associated haplotypes. The haplotypes are ordered by(More)
The recently developed Bayesian multinomial calibration model, Bummer, was applied to the fossil chironomid assemblages of a lake situated at the present tree-line in northern Fennoscandia to reconstruct patterns in Holocene climates. The results are compared with inferences derived using non-Bayesian weighted averaging partial least-squares regression. The(More)
Segmentation is a popular technique for discovering structure in time series data. We address the largely open problem of estimating the number of segments that can be reliably discovered. We introduce a novel method for the problem, called Pete. Pete is based on permutation testing. The problem is an instance of model (dimension) selection. The proposed(More)
We describe a new method for linkage disequilibrium mapping, Haplotype Pattern Mining (HPM). The method is based on discovering recurrent patterns, inspired by data mining methods. We define a class of useful haplotype patterns in genetic case-control data, and give an algorithm for finding disease-associated haplotypes. The haplotypes are ordered by their(More)
Segmentation analysis addresses the following data analysis problem: given a time series, find a partitioning of the sequence to segments that are internally homogenous with respect to the desired pattern language and cost function. We will consider applications of segmentation analysis towards analysis of paleoecological spatio-temporal time series data.(More)
In order to compute a reliable score for a Bayesian network structure B with respect to its conditional independence assertions one should take into account the size of B's equivalence class s(B). So far, no one has explicitly presented an algorithm that would calculate s(B) eeectively. We introduce an algorithm called BE1 that seems to work fast on most of(More)
  • 1