Skip to content
Permalink
Browse files

Fix probability-weighted distance for S figure.

  • Loading branch information...
evogytis committed Mar 19, 2019
1 parent 6cd471f commit b67c15d5ee2ccfda2af74391d6f39487188303f7
Showing with 1 addition and 1 deletion.
  1. +1 −1 genomic-horizon.tex
@@ -537,7 +537,7 @@ \section*{Acknowledgements}
\includegraphics[width=0.75\textwidth]{supp_figures/sfigX_treetimeLocations.png}
\caption{\textbf{Maximum likelihood inference of masked sequence location from genomes (left) and GP sequences (right) via a CTMC model implemented in TreeTime.}
Unlike the Bayesian GLM where probabilities are spread across different locations, the CTMC gives categorical support to one location.
Genomes still perform better in terms of correct guess (0.432 probability that best guess location is true location for genomes versus 0.259 for GP), cross entropy (12012.800 nats for genome versus 24397.109 nats for GP) and mean probability-weighted great circle distance between true location population centroid and estimated location population centroid (2.304 km for genome versus 3.287 km for GP).
Genomes still perform better in terms of correct guess (0.432 probability that best guess location is true location for genomes versus 0.259 for GP), cross entropy (12012.800 nats for genome versus 24397.109 nats for GP) and mean probability-weighted great circle distance between true location population centroid and estimated location population centroid (87.568 km for genome versus 124.909 km for GP).
}
\label{TTlocations}
\end{figure}

0 comments on commit b67c15d

Please sign in to comment.
You can’t perform that action at this time.