ndwarshuis
/
phd_thesis
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

ENH move IL2 stuff to its own section

This commit is contained in:
Nathan Dwarshuis 2021-07-29 13:14:40 -04:00
parent 54fb855b19
commit 1c65546e41
1 changed files with 14 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

23
tex/thesis.tex
View File

@ -2168,7 +2168,7 @@ Venn diagram from the venn R package.
\section{results} \section{results}
\subsection{DOE shows optimal conditions for expanded potent T cells} \subsection{T cells can be grown on DMSs with lower IL2 concentrations}
% TODO this plots proportions look dumb % TODO this plots proportions look dumb
\begin{figure*}[ht!] \begin{figure*}[ht!]
@ -2194,6 +2194,10 @@ Venn diagram from the venn R package.
\label{fig:il2_mod} \label{fig:il2_mod}
\end{figure*} \end{figure*}
% TODO this is not consistent with the next section since the responses are
% different
\subsection{DOE shows optimal conditions for expanded potent T cells}
% TODO not all of these were actually use, explain why by either adding columns % TODO not all of these were actually use, explain why by either adding columns
% or marking with an asterisk % or marking with an asterisk
\begin{table}[!h] \centering \begin{table}[!h] \centering
@ -2284,14 +2288,15 @@ process (Fig.1d-e).
\end{table} \end{table}
SR models achieved the highest predictive performance (R2>93\%) when using SR models achieved the highest predictive performance (R2>93\%) when using
multi-omics predictors for all endpoint responses (\cref{tab:mod_results}). SR achieved R2>98\% multi-omics predictors for all endpoint responses (\cref{tab:mod_results}). SR
while GBM tree-based ensembles showed leave-one-out cross-validated R2 (LOO-R2) achieved R2>98\% while GBM tree-based ensembles showed leave-one-out
>95\% for CD4+ and CD4+/CD8+ TN+TCM responses. Similarly, LASSO, PLSR, and SVM cross-validated R2 (LOO-R2) >95\% for CD4+ and CD4+/CD8+ TN+TCM responses.
methods showed consistent high LOO-R2, 92.9\%, 99.7\%, and 90.5\%, respectively, Similarly, LASSO, PLSR, and SVM methods showed consistent high LOO-R2, 92.9\%,
to predict the CD4+/CD8+ TN+TCM. Yet, about 10\% reduction in LOO-R2, 99.7\%, and 90.5\%, respectively, to predict the CD4+/CD8+ TN+TCM. Yet, about
72.5\%-81.7\%, was observed for CD4+ TN+TCM with these three methods. Lastly, SR 10\% reduction in LOO-R2, 72.5\%-81.7\%, was observed for CD4+ TN+TCM with these
and PLSR achieved R2>90\% while other ML methods exhibited exceedingly variable three methods. Lastly, SR and PLSR achieved R2>90\% while other ML methods
LOO-R2 (0.3\%,RF-51.5\%,LASSO) for CD8+ TN+TCM cells. exhibited exceedingly variable LOO-R2 (0.3\%,RF-51.5\%,LASSO) for CD8+ TN+TCM
cells.
% FIGURE the CD4/CD8 model results using SR % FIGURE the CD4/CD8 model results using SR