Search Results for "em algorithm"

This project implements regression models based on expectation maximization (EM) algorithms in case of missing data

karkinos is tumor genotyper which detects single nucleotide variation (SNV), integer copy number variation (CNV) and calculates tumor cellularity from tumor-normal paired sequencing data. Accurate CNV calling is achieved using continuous wavelet analysis and multi-state HMM, while SNV call is adjusted by tumor cellularity and filtered by heuristic filtering algorithm and Fisher Test. Also, Noise calls in low depth region are removed using EM algorithm.

MITSU is an algorithm for discovery of transcription factor binding site (TFBS) motifs. It is based on the stochastic EM (sEM) algorithm, which overcomes some of the limitations of deterministic EM-based algorithms for motif discovery. Unlike previous sEM algorithms for motif discovery, MITSU is unconstrained with regard to the distribution of motif occurrences within the input dataset. MITSU also has the ability to automatically determine the most likely motif width by incorporating...

Stochastic discriminative EM (sdEM) is an online-EM-type algorithm for discriminative training of probabilistic generative models belonging to the natural exponential family. In this work, we introduce and justify this algorithm as a stochastic natural gradient descent method, i.e. a method which accounts for the information geometry in the parameter space of the statistical model. We show how this learning algorithm can be used to train probabilistic generative models by minimizing different...

... parameters by analysing next generation sequencing data. We offer an implementation of the EM algorithm to find maximum a posteriori estimates of the model parameters and a method to estimate the distribution of viral strains in the quasispecies. The model is validated on simulated data, showing the advantage of explicitly taking the recombination process into account, and tested by applying to reads obtained from experimental HIV samples.

This little software is the realization of EM algorithm in the application of tossiing the coin, which is described in the paper of Michael Collins in 1997.

EMPEPA is software that permits to find the most likely rates of a PEPA model according to a set of sample executions by using the EM algorithm. It uses the GNU Scientific Library (GSL).