<!DOCTYPE html>
<html>
<head>
<meta http-equiv="Content-Type" content="text/html;charset=utf-8">
<meta http-equiv="X-UA-Compatible" content="IE=edge">
<meta name="viewport"
content="user-scalable=no, width=device-width, initial-scale=1, maximum-scale=1">
<title>phantasus - Tutorial</title>
<link rel="stylesheet" href="css/phantasus-latest.min.css">
<script type="text/javascript" src="js/phantasus-external-latest.min.js"></script>
<script src="js/phantasus-latest.min.js"></script>
<link rel="shortcut icon" href="favicon.ico" type="image/x-icon">
</head>
<body>
<div class="container-fluid">
<div class="row">
<div class="col-xs-12">
<h4>
Tutorial: <a target="_blank"
href="http://www.broadinstitute.org/achilles">Project Achilles</a>
and the <a target="_blank"
href="http://www.broadinstitute.org/ccle/home">Cancer Cell
Line Encyclopedia (CCLE)</a> datasets.
</h4>
</div>
</div>
<div id="header" class="row">
<div class="col-xs-12">
<input type="button" class="btn btn-default btn-sm"
value="Expand All" name="expand"> <input type="button"
class="btn btn-default btn-sm" value="Collapse All" name="collapse">
</div>
</div>
<div class="row">
<div class="col-xs-12">
<a data-toggle="collapse"
style="font-size: 1.2em; display: block; font-weight: 800;">Open
Project Achilles Gene Essentiality Scores</a>
<p class="collapse">Open the Project Achilles gene essentiality
scores under the preloaded datasets section. By default, values in
the heat map are mapped to colors using the minimum and maximum of
each row independently.</p>
<a data-toggle="collapse"
style="font-size: 1.2em; display: block; font-weight: 800;">Median
Absolute Deviation (MAD)</a>
<p class="collapse">
Click the tools icon
<button type="button"
class="btn btn-default btn-xs dropdown-toggle">
<span class="fa fa-wrench"></span> <span class="fa fa-caret-down"></span>
</button>
and select "Create Calculated Annotation". <br /> Enter "MAD" for
the "Annotation name". <br /> Enter "MAD()" for "Formula" and
click "OK". <br /> A new row annotation named "MAD" will appear to
the right of the heat map.
</p>
<a data-toggle="collapse"
style="font-size: 1.2em; display: block; font-weight: 800;">Filter
Rows</a>
<p class="collapse">
Show only the top 500 most variable genes by the median absolute
deviation <br /> Click the filter icon
<button type="button"
class="btn btn-default btn-xs dropdown-toggle">
<span class="fa fa-filter"></span> <span class="fa fa-caret-down"></span>
</button>
. <br />Click the "Add" button and set the field name to "MAD". <br />Click
"Switch to top N filter" and enter "500" for "N". Note that 523
rows are now shown because of ties in the median absolute
deviation.
</p>
<a data-toggle="collapse"
style="font-size: 1.2em; display: block; font-weight: 800;">Hierarchical
Clustering</a>
<p class="collapse">
Click the tools icon
<button type="button"
class="btn btn-default btn-xs dropdown-toggle">
<span class="fa fa-wrench"></span> <span class="fa fa-caret-down"></span>
</button>
and select "Hierarchical Clustering". Change "Cluster" to "Rows and
columns". Click "OK" to run the analysis. <br /> Do haematopoietic
cell lines cluster together? Hover over the primary_site
annotations to highlight cell lines from the same primary site. You
can dynamically cut the dendrogram by dragging the dashed line at
the top of the dendrogram. <br />Hierarchical clustering
recursively merges objects based on their pair-wise distance.
Objects closest together are merged first, objects furthest apart
are merged last. The result is a tree structure, referred to as a
dendogram, where the leaf nodes represent the original items and
internal (higher) nodes represent the merges that occurred. Click <a
target="_black"
href="http://www.mathworks.com/help/stats/hierarchical-clustering.html">here</a>
for a more detailed description of the hierarchical clustering
algorithm and <a target="_blank"
href="http://support.minitab.com/en-us/minitab-express/1/help-and-how-to/modeling-statistics/regression/supporting-topics/basics/a-comparison-of-the-pearson-and-spearman-correlation-methods/">here</a>
for a comparison of the Pearson and Spearman correlation methods.
</p>
<a data-toggle="collapse"
style="font-size: 1.2em; display: block; font-weight: 800;">Compress
Heat Map</a>
<p class="collapse">
Click the fit to window icon
<button type="button" class="btn btn-default btn-xs">
<span class="fa fa-compress"></span>
</button>
to fit compress the heat map. Click
<button type="button" class="btn btn-default btn-xs"
title="Reset Zoom" name="resetZoom">100%</button>
to return to the normal heat map size.
</p>
<a data-toggle="collapse"
style="font-size: 1.2em; display: block; font-weight: 800;">Remove
Row Filter</a>
<p class="collapse">
Click the filter icon
<button type="button"
class="btn btn-default btn-xs dropdown-toggle">
<span class="fa fa-filter"></span> <span class="fa fa-caret-down"></span>
</button>
and delete the median absolute deviation filter to show all rows.
</p>
<a data-toggle="collapse"
style="font-size: 1.2em; display: block; font-weight: 800;">Remove
Dendrograms</a>
<p class="collapse">Right-click on the column dendrogram and
select "Delete". Do the same for the row dendrogram.</p>
<a data-toggle="collapse"
style="font-size: 1.2em; display: block; font-weight: 800;">Sort
Rows</a>
<p class="collapse">
Click the MAD row annotation header in the heat map once to sort
the heat map in ascending order by the median absolute deviation.
Click it again to sort the heat map in descending order. Note that
you can shift-click to sort multiple columns simultaneously.
Alternatively, you can click
<button title="Sort" name="sort" type="button"
class="btn btn-default btn-xs">
<span class="fa fa-sort-alpha-asc"></span>
</button>
to open a sort dialog.
</p>
<a data-toggle="collapse"
style="font-size: 1.2em; display: block; font-weight: 800;">Search
Results To Top</a>
<p class="collapse">
Enter "BRAF" in the row search box. <br /> Click
<button type="button" class="btn btn-default btn-xs">
<i class="fa fa-level-up"></i>
</button>
to bring the matches to the top of the heat map.
</p>
<a data-toggle="collapse"
style="font-size: 1.2em; display: block; font-weight: 800;">Visual
Enrichment</a>
<p class="collapse">
Double-click on BRAF row to sort by dependency score. <br />
Search columns for "skin". <br /> Do skin cell lines seem to be
more dependent on BRAF than cell lines from other lineages? <br />
You can optionally limit your search to within a field by typing
the field name followed by a colon ":" and then the term you are
looking for.
</p>
<a data-toggle="collapse"
style="font-size: 1.2em; display: block; font-weight: 800;">Nearest
Neighbors</a>
<p class="collapse">
Click the tools icon
<button type="button"
class="btn btn-default btn-xs dropdown-toggle">
<span class="fa fa-wrench"></span> <span class="fa fa-caret-down"></span>
</button>
and select "Nearest Neighbors". Click "OK" to run the analysis. <br />
A new row annotation named "Pearson correlation" will appear. <br />
What genes correlate with BRAF?
</p>
<a data-toggle="collapse"
style="font-size: 1.2em; display: block; font-weight: 800;">Scatter
Plot Matrix</a>
<p class="collapse">
Select the first three rows in order to select the top 3 most
correlated genes with BRAF and all samples. Click the chart icon
<button type="button" class="btn btn-default btn-xs">
<span class="fa fa-line-chart"></span>
</button>
. Change the chart type to "row scatter" to assess the
relationships between BRAF, FXR2, and ZNF781 simultaneously.
</p>
<a data-toggle="collapse"
style="font-size: 1.2em; display: block; font-weight: 800;">Append
Mutations</a>
<p class="collapse">
Click
<button type="button" class="btn btn-default btn-xs">
<span class="fa fa-folder-open-o"></span>
</button>
to bring up the file open window. <br /> Check "Append rows to
current dataset" and select the file CCLE Mutations dataset. Change
"Current dataset annotation name" and "New dataset annotation name"
to "id" to match the ids in the existing dataset with the id in the
mutational dataset. Scroll down to see the mutational data.
</p>
<!-- <a data-toggle="collapse" -->
<!-- style="font-size: 1.2em; display: block; font-weight: 800;">Conditional -->
<!-- Heat Map Color Schemes</a> -->
<!-- <p class="collapse"> -->
<!-- Select -->
<!-- <button type="button" class="btn btn-default btn-xs"> -->
<!-- <span class="fa fa-cog"></span> -->
<!-- </button> -->
<!-- to open the options window. <br /> Change "Color by" to "Source" <br /> -->
<!-- Change "Color by value" to -->
<!-- "CCLE_hybrid_capture1650_hg19_NoCommonSNPs_NoNeutralVariants_CDS_2012.05.07" -->
<!-- <br /> Change "Predefined color schemes" to "MAF". -->
<!-- </p> -->
<a data-toggle="collapse"
style="font-size: 1.2em; display: block; font-weight: 800;">Toggle
Mutation Visibility</a>
<p class="collapse">
Click
<button title="Color Key" type="button"
class="btn btn-default btn-xs" aria-expanded="false">
<span class="fa fa-key"></span>
</button>
to open the color key. <br /> Click the "Synonymous" check box to
visually hide synonymous mutations (note the matrix values are left
unchanged).
</p>
<!-- <a data-toggle="collapse" -->
<!-- style="font-size: 1.2em; display: block; font-weight: 800;">Group -->
<!-- Rows</a> -->
<!-- <p class="collapse"> -->
<!-- Click -->
<!-- <button title="Sort" name="sort" type="button" -->
<!-- class="btn btn-default btn-xs"> -->
<!-- <span class="fa fa-sort-alpha-asc"></span> -->
<!-- </button> -->
<!-- to open the sort and group by dialog. Sort by "Source". Next, -->
<!-- select "Source" under the "Group by" field. Grouping adds -->
<!-- whitespace between different values of the group by fields. -->
<!-- </p> -->
<a data-toggle="collapse"
style="font-size: 1.2em; display: block; font-weight: 800;">Annotate
Selection</a>
<p class="collapse">
Enter "BRAF" in the row search box. <br /> Click
<button type="button" class="btn btn-default btn-xs">
<i class="fa fa-level-up"></i>
</button>
to bring the matches to the top of the heat map. Double-click to
sort the heat map by the BRAF dependency score. Select the seven
cell lines that have a dependency score less than or equal to -2.
Right-click and select "Annotate Selection". Enter "BRAF_sensitive"
for "Annotation name" and "y" for "Annotation value".
</p>
<a data-toggle="collapse"
style="font-size: 1.2em; display: block; font-weight: 800;">New
Heat Map</a>
<p class="collapse">
Enter "CCLE" in the row search box to select the CCLE mutation
data. <br />Right-click on the columns and select "Clear
Selection" to clear the column selection. <br />Click the tools
icon
<button type="button"
class="btn btn-default btn-xs dropdown-toggle">
<span class="fa fa-wrench"></span> <span class="fa fa-caret-down"></span>
</button>
and select "New Heat Map". A new tab will open that contains the
selected subset of the heat map.
</p>
<a data-toggle="collapse"
style="font-size: 1.2em; display: block; font-weight: 800;">Fisher
Exact Test</a>
<p class="collapse">
Click the tools icon
<button type="button"
class="btn btn-default btn-xs dropdown-toggle">
<span class="fa fa-wrench"></span> <span class="fa fa-caret-down"></span>
</button>
and select "Marker Selection". Change "Metric" to "Fisher Exact
Test". Change "Field" to "BRAF_sensitive". Set "Class a" to
"BRAF_sensitive" and "Class b" to "". Set "Grouping value" to "4"
to group missense, splice site, frame shift, and nonsense mutations
together and no mutation, synonymous, in-frame indels, and other
non-synonymous mutations together (Click
<button title="Color Key" type="button"
class="btn btn-default btn-xs" aria-expanded="false">
<span class="fa fa-key"></span>
</button>
to see the numerical values used to encode mutation type). Click
"OK" to run the analysis. <br /> A new row annotation named
"Fisher Exact Test" will appear. <br /> What mutations distinguish
BRAF sensitive cell lines from the other cell lines?
</p>
</div>
</div>
<div class="row"></div>
<div class="row"></div>
<div class="row"></div>
<div class="row"></div>
<div class="row"></div>
<div class="row"></div>
</div>
<script type="text/javascript">
var links = $(document.body).find('[data-toggle=collapse]');
var collapse = $(document.body).find('.collapse');
for (var i = 0; i < collapse.length; i++) {
var $c = $(collapse[i]);
var $l = $(links[i]);
$l.html((i + 1) + '. ' + $l.html());
$c.attr('id', i);
$l.attr('href', '#' + i);
}
$('#header').find('[name=expand]').on('click', function() {
$(document.body).find('.collapse').collapse('show');
});
$('#header').find('[name=collapse]').on('click', function() {
$(document.body).find('.collapse').collapse('hide');
});
</script>
<script>
phantasus.Util.loadTrackingCode();
</script>
</body>
</html>
