Skip to main content
Premium Trial:

Request an Annual Quote

This Week in PLoS: Oct 14, 2008

David Gilbert and his colleagues investigated how replication domains in embryonic stem cells reorganize during differentiation. They created genome-wide replication-timing maps in mouse embryonic stem cells. After the mouse cells differentiated into neural precursor cells, 20 percent of the genome then had different replication timing -- the timing domains consolidated. "It is possible that establishing replication maps will generate a database of chromosome segments that undergo large changes in organization during differentiation as well as changes in the locations and polarities of replication forks that must travel long distances between replication domains," the authors write in PLoS Biology.

Researchers from Imperial College London studied the noise found in gene expression, particularly that due to a "burst" of protein production activity due to a transcription factor binding. Using queuing theory, they created a theoretical probability distribution of the bursts for a few gene expression models and found that burst distribution is always the same shape. The researchers add that with more data, such as protein equilibrium distributions, more parameters can be inferred. Their work is in last week's PLoS Computational Biology.

Hong-Wen Deng and his colleagues compared population-based association methods by simulating stratified populations based on real data from the HapMap project. They compared the relative power, type I error rates, accuracy and positive prediction value of case-control studies, structured association, genomic control, and principal components analysis studies. "Our study intends to provide a practical guideline for researchers to select proper study methods and make appropriate inference of the results in population-based association studies," write the authors in PLoS One.

Also in PLoS One, John Woolliams' team reports deriving formulae to predict the accuracy of a genetic disease risk from a population or case-control study, for both continuous and dichotomous phenotypes. They tested these predictions in simulations and saw that they fit well.