figshare
Browse
Figure_4.tif (283.6 kB)

Results from the pattern analysis.

Download (0 kB)
figure
posted on 2013-05-09, 00:15 authored by Richard A. White, Jørgen V. Bjørnholt, Donna D. Baird, Tore Midtvedt, Jennifer R. Harris, Marcello Pagano, Winston Hide, Knut Rudi, Birgitte Moen, Nina Iszatt, Shyamal D. Peddada, Merete Eggesbø

The exposure pattern is represented by four characters, constructed from “”, “”, and “X”, which represent detection, non-detection, and irrelevance, respectively, for the four time points of the analysis (days 4, 10, 30, and 120). The black points and lines represent estimated means and 95% confidence intervals for patterns that were found to be significantly close to zero at an false discovery rate of 5%. The crude contrasts (i.e. if “XX” was significant, the crude contrast would be “XX”) that were significantly different to zero at have their estimated means and 95% confidence intervals displayed in red. For the testing of the difference of the means of the two patterns, significant results (at ) is indicated by , otherwise significance is . Vertical lines are displayed at and , representing the boundaries of expected growth.

History

Usage metrics

    PLOS Computational Biology

    Licence

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC