this is for holding javascript data
Charith Bhagya Karunarathna edited untitled.tex
almost 8 years ago
Commit id: c859936da62d817070fdee53c8b92d8759f387d3
deletions | additions
diff --git a/untitled.tex b/untitled.tex
index 3240a0f..ecb5306 100644
--- a/untitled.tex
+++ b/untitled.tex
...
Elastic-net \cite{Zou_2005} is a hybrid regularization and variable selection method that linearly combines the $L1$, and $L2$ regularization penalties of Lasso \cite{Tibshirani_2011}, and Ridge \cite{Cessie_1992} methods in multivariate regression. This combination of Lasso and Ridge penalties provides a more precise prediction than using multiple regression, when SNVs are in high linkage disequilibrium. We here used elastic-net to select disease-causing SNVs because elastic-net method is particular useful when number of predictors exceeds the number of observations. To validate these selected disease-causing SNVs via elastic-net, we employed replication study using $100$ bootstrap samples. We then evaluated bootstrap variable (SNV) inclusion probability which is a frequentist analog of Bayesian posterior inclusion probability from CAVIABF.
\subsection{Tree-Based method}
We here evaluated two tree based methods: Blossoc ((BLOck aSSOCiation, \cite{Mailund_2006}), and Mantel test based on rank of coalescent events \cite{Burkett_2013}.
Blossoc is a fast method to localize the disease-causing varinats by reconstructing genealogical trees at each SNV site.