Commit 7328d37f authored by waldispuhl's avatar waldispuhl
Browse files

revision JW

parent 7ad05b4e
\documentclass[letterpaper,12pt]{article}
\usepackage[margin=1in]{geometry}
\usepackage{epsf}
\usepackage{amssymb} % AMS symbols
\usepackage{latexsym} % Old Latex symbols
\usepackage{epsfig}
\usepackage{graphics}
\usepackage{xspace}
\usepackage{url}
\usepackage{todonotes}
\renewcommand{\em}{\it}
\parindent=0in
\renewcommand{\thepage}{ } \pagestyle{empty}
\begin{document}
\includegraphics[width=0.3\linewidth]{mcgill_logo}\\
\hrule
\vspace{1em}
\hspace{11cm}
\parbox{2.5in}
{\small J\'er\^ome Waldisp\"{u}hl, Ph.D.\\
School of Computer Science\\
McGill Centre for Bioinformatics\\
McGill University\\
{\tt jeromew@cs.mcgill.ca} \\
\today}
\vskip 0.3in
Dear Dr. Alan Kimmel,
\vskip 0.3in
We thank you for giving us an opportunity to improve our manuscript. We do agree with the reviewer that additional validations will improve the impact and significance of our work.\\
We thank the reviewer for pointing at us novel mutate-and-map experiments that have been released after the initial submission of our manuscript. In this revision, we include all new mutate-and-map experiments for which we could (i) identify evolutionary conserved intermolecular interactions, (ii) find an experimentally determined structure in the PDB, and (iii) obtain an alignment from the Rfam family. We provide below a list of all experiments included in our updated benchmark, and a justification for the ones that we could not included inside.
\section*{Molecules included in the benchmark}
\subsubsection*{5S rRNA}
This molecule was already included in the benchmark of our original manuscript.
\subsubsection*{c-di-GMP Riboswitch}
This molecule was already included in the benchmark of our original manuscript.
\subsubsection*{Adenine Riboswitch (New)}
Initially, we reported the molecule in the supplementary material because only one nucleotide was found to interact with another molecule. However, we discovered that the automated annotation, originally done with a Pymol script, was incomplete.\\
We manually revisited the PDB file associated with this molecule (1Y26), and identified 9 additional interacting positions. In light of this finding, we decided to move this molecule into the main body of our manuscript.\\
The mutate-and-map repository has 3 different experiments on this molecule (named adenine\_2, adenine\_3, and adenine\_4, following the mutate-and-map annotation). However, only the first two experiments were conducted with the ligand, the last one does not have it. Strikingly, our methods perform well on the two first mutate-and-map experiments (with the ligand), but fail to identify a clear signal on the third one (without the ligand). This observation brings additional support to our approach. Indeed, in the absence of the ligand, the structure differs from the ligand-bound conformation. The mutate-and-map mutations altering the structure of the non-bound conformation are not necessarily associated with changes of the ligand-bound structure, and the evolutionary information calculated with our algorithms cannot be related to binding sites of the ligand-bound structure.
\subsubsection*{tRNA phenylalanine (New)}
The PDB contains two experimental structures of this RNA in the yeast 80S ribosome- tRNA complexes (PDB identifier 3J78), and Rfam provides a single alignment for \emph{all} tRNAs (RF00005). We clearly identified 2 conserved regions of interactions in the literature. The anticodon and the T-$\Psi$-C-G motif binding the 5S RNA in the 50S ribosomal subunit. We thus included these two sites in the revised benchmark. We present and discuss the results separately.\\
A couple of other nucleotides are also eventually implicated in interactions with other molecules (positions 1, 19 and 73-76). However, these positions seem to be weakly evolutionary conserved. In fact, interacting positions vary between the two copies of the tRNA in the PDB structure of the 80S ribosome. Our methods are not designed to work in such cases. For this reason, we did not included these interaction sites in our benchmark. Although, for the sake of transparency, we included the results on these sites in the supplementary material.
\subsubsection*{Cobalamin riboswitch (New)}
Also referred as RNAPuzzle 6 in the mutate-and-map repository, this molecule is associated to the RNA family RF00174. The PDB has an experimental structure (4GXY),which has been crystallized in the presence of magnesium and iridium. We ignored these molecules in our experiments and focused on the ligand binding pocket.\\
Interestingly,
\section*{Molecules not included in the benchmark}
\subsubsection*{Glycine Riboswitch}
We tried to integrate the Glycosine riboswitch in our benchmark. The latter is associated with the PDB structure 3P49 and Rfam family RF00504.\\
Unfortunately, we realized after the alignment of the PDB sequence with the Rfam alignment that the target sequence contains an artificial sequence of a stem loop used to bind protein stabilizing the whole RNA structure. The large discrepancy between the PDB sequence and mutate-and-map sequences prevented us to use this molecules in our benchmark.\\
Nonetheless, in the revised version of this paper, we discuss this case and provide all results in the supplementary material.\\
\subsection*{Other molecules}
The 16S rRNA Four-Way Junction has not been included in our new benchmark because we have not been able to identify evolutionary conserved intermolecular interactions.\\
The SAM riboswitch (RNAPuzzle 8) has not been included in the new benchmark because the sequence similarity of the mutate-and-map sequence with Rfam alignment (RF00162) is below 50\%. Other related experimental structures present in the PDB have similar issues.\\
The following molecules were not included in our new benchmark because we have not been able to identify an alignment in the Rfam database:
\begin{itemize}
\setlength{\itemsep}{-1pt}
\item P4-P6 domain, Tetrahymena ribozyme.
\item M-stableRNA.
\item Hox A9 mRNA 5' UTR.
\item MedLoop.
\item Class I Ligase.
\end{itemize}
The following molecules were not integrated in our benchmark because these are artificial sequences:
\begin{itemize}
\setlength{\itemsep}{-1pt}
\item Hobartner bistable switch
\item X20/H20
\item Tebowned
\end{itemize}
\end{document}
This diff is collapsed.
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment