Skip to content

Instantly share code, notes, and snippets.

@nielshanson
Last active August 29, 2015 14:17
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save nielshanson/ffc776183b15d8652275 to your computer and use it in GitHub Desktop.
Save nielshanson/ffc776183b15d8652275 to your computer and use it in GitHub Desktop.
Heatmap

About the plot

A two-variable heatmap with two-sided marginals and mouseover. Heatmaps are used all over the pace, but in the field of Metagenomics Bubble Plots are popular for displaying the functional information of a microbial community (a loose convention). This work was heavily inspired heavily by the Les Misérable Co-occurence example designed by Mike Bostock. Marginals distributions are automatically calculated on the input csv-matrix.

Being displayed is a functioal screen of the Carbohydrate-Active enZYmes Database (CAZy), a collection of genes and proteins associated with the degradation of various complex carbohydrate sugars. Of particular interest from a material and chemical development perspective is the degradation of Lignin and Cellulose, two of the most abundant polymers on Earth.

Data has been graciously provided by Mr. Keith Mewis , UBC Genome Sciences and Technology Ph.D student from the Steven J. Hallam Laboratory. Slightly modified versions of this code has been in the following publications:

Usage

A simple comma-separated value file with row-names and column-names is sufficient.

SVG Crowbar is a great tool for extracting SVG files from d3 plots for downstream purposes.

body {
background-color: white;
color: black;
font-family: Arial,Helvetica Neue,Helvetica,sans-serif;
}
div.bar {
display: inline-block;
width: 20px;
height: 75px; /* We'll override this later */
background-color: teal;
}
#table {
width: 800px;
margin-left: auto ;
margin-right: auto ;
}
.background {
fill: #eee;
}
.axis path,
.axis line {
fill: none;
stroke: black;
shape-rendering: crispEdges;
}
.axis text {
font-family: sans-serif;
font-size: 11px;
}
/* Hover info */
.hoverinfo {
font-size: 9px;
padding: 4px;
border-radius: 1px;
background-color: #FFF;
box-shadow: 1px 1px 5px #CCC;
font-size: 12px;
border: 1px solid #CCC;
}
d3.text("https://cdn.rawgit.com/nielshanson/d3/master/data/ReducedGHMining_trans.csv", function(datasetText) {
var parsedCSV = d3.csv.parseRows(datasetText);
// add tooltip
var tooltip = d3.select("body")
.append("div")
.style("position", "absolute")
.style("z-index", "10")
.attr("class", "hoverinfo")
.style("visibility", "hidden");
// parse out x and y axis labels
var xlab = parsedCSV[0].slice(1,parsedCSV[0].length);
var ylab = [];
for(var i = 1; i < parsedCSV.length; i++){
var temp = parsedCSV[i];
ylab = ylab.concat(temp[0]);
}
// parse the datamatrix
var data_matrix = [];
for(var i = 1; i < parsedCSV.length; i++){
var temp = parsedCSV[i].slice(1,parsedCSV[i].length );
data_matrix = data_matrix.concat([temp]);
}
// find marginal sums
var sum_rows = []; // sum of the rows
for(var i = 0; i < data_matrix.length; i++){
var temp_sum = 0;
for(var j = 0; j < data_matrix[i].length; j++){
temp_sum = parseFloat(temp_sum) + parseFloat(data_matrix[i][j]);
}
sum_rows = sum_rows.concat([temp_sum]);
}
var sum_col = []; // sum of the columns
for(var i = 0; i < data_matrix[0].length; i++){
var temp_sum = 0;
for(var j = 0; j < data_matrix.length; j++){
temp_sum = parseFloat(temp_sum) + parseFloat(data_matrix[j][i]);
}
sum_col = sum_col.concat([temp_sum]);
}
function colorPicker(d) {
if( d <= 0 ){
return "rgb(255, 247, 251)"; // 0
} else if (d > 0 && d<= 2){
return "rgb(236, 226, 240)"; // 1
} else if (d > 2 && d<= 4) {
return "rgb(208, 209, 230)"; // 2
} else if (d > 4 && d<= 6) {
return "rgb(166, 189, 219)"; // 3
} else if (d > 6 && d<= 8){
return "rgb(103, 169, 207)"; // 4
} else if (d > 8 && d<= 10){
return "rgb(54, 144, 192)"; // 5
} else if (d > 10 && d<= 12){
return "rgb(2, 129, 138)"; // 6
} else if (d > 12 && d<= 14) {
return "rgb(1, 108, 89)"; // 7
} else if (d > 14){
return "rgb(1, 70, 54)"; // 7+
} else {
return "pink"; // number not in range
}
}
var margin = {top: 160, right: 160, bottom: 80, left: 160},
width = data_matrix[1].length*12;
height = data_matrix.length*12;
var svg = d3.select("#table").append("svg");
svg.attr("width", width + margin.left + margin.right)
.attr("height", height + margin.top + margin.bottom)
.style("border-collapse", "collapse")
.append("g")
.attr("transform", "translate(" + margin.left + "," + margin.top + ")")
.selectAll("g")
.data(data_matrix)
.enter()
.append("g")
.attr("class", "row")
.attr("transform", function(d, i){ return "translate(0," + (i*12) + ")"; })
.selectAll("rect")
.data(function(d){return d;})
.enter().append("rect")
.attr("width", "10px")
.attr("height", "10px")
.attr("fill", function(d) {
return colorPicker(d);
})
.attr("x", function(d,i) {
return (i * 12);
})
.on("mouseover", function(d,i){
d3.select(this).attr("fill", "#F77E1C")
tooltip.html(d);
return tooltip.style("visibility", "visible");
})
.on("mouseout", function(){d3.select(this).attr("fill", function(d) {
tooltip.style("visibility", "hidden");
return colorPicker(d);
})})
.on("mousemove", function(){
return tooltip.style("top",(d3.event.pageY+10)+"px").style("left",(d3.event.pageX+10)+"px");
})
.on("mousedown", function(){d3.select(this).attr("fill", "red")})
.on("mouseup", function(){d3.select(this).attr("fill", "#F77E1C")})
//.attr("onclick", "javascript:showBlast('MMGA_0001')");
.attr("alt", function(d) {
return d;
})
// axis
var xScale = d3.scale.ordinal()
.domain(xlab)
.rangePoints([0+5, width-5]);
var xAxis = d3.svg.axis()
.scale(xScale)
.orient("left")
var yScale = d3.scale.ordinal()
.domain(ylab)
.rangePoints([0+4, height-4]);
var yAxis = d3.svg.axis()
.scale(yScale)
.orient("left")
svg.append("g")
.attr("class", "axis")
.attr("transform", "translate("+ margin.left + "," + (margin.top + height) + ") rotate(270)")
.call(xAxis);
svg.append("g")
.attr("class", "axis")
.attr("transform", "translate("+ (margin.left -2) + "," + margin.top + ") rotate(0)")
.call(yAxis);
//
var col_hist_size = 0.5* margin.top;
var col_hist_scale = d3.scale.linear()
.domain([0, d3.max(sum_col)])
.range([0, col_hist_size]);
svg.append("g")
//.attr("transform", "translate(" + margin.left + "," + margin.top + ")")
.attr("transform", "translate(" + margin.left + ",0)")
.selectAll("rect")
.data(sum_col)
.enter()
.append("rect")
.attr("x", function(d, i) {
return i * 12; //Bar width of 12 plus 1 for padding
})
.attr("y", function(d) {
return margin.top - col_hist_scale(d) - 5; // spacing of 5
})
.attr("width", 10)
.attr("fill","teal")
.attr("height", function(d) {
return col_hist_scale(d);
});
svg.append("g")
.attr("transform", "translate(" + margin.left + ",0)")
.selectAll("text")
.data(sum_col)
.enter()
.append("text")
.text(function(d) {
return d;
})
.attr("x", function(d, i) {
return i * 12 + 2 + 3;
})
.attr("y", function(d) {
return margin.top - col_hist_scale(d) - 5 - 3;
})
.attr("font-family", "sans-serif")
.attr("font-size", "8px")
.attr("fill", "black")
.attr("text-anchor", "middle");
svg.append("g").append("line")
.attr("x1", margin.left).attr("x2", width + margin.right - 2).attr("y1", margin.top - 3).attr("y2", margin.top - 3).attr("stroke","black");
// row histogram
var row_hist_size = 0.8* margin.right;
var row_hist_scale = d3.scale.linear()
.domain([0, d3.max(sum_rows)])
.range([0, row_hist_size]);
svg.append("g")
.attr("transform", "translate(" + (width + margin.right) + "," + margin.top + ")")
.selectAll("rect")
.data(sum_rows)
.enter()
.append("rect")
.attr("x", function(d, i) {
return 0; // spacing of 5
})
.attr("y", function(d, i) {
return i * 12; //Bar width of 12 plus 1 for padding
})
.attr("width", function(d) {
return row_hist_scale(d);
})
.attr("fill","rgb(54, 144, 192)")
.attr("height", 10);
svg.append("g")
.attr("transform", "translate(" + (width + margin.right) + "," + margin.top + ")")
.selectAll("text")
.data(sum_rows)
.enter()
.append("text")
.text(function(d) {
return d;
})
.attr("x", function(d, i) {
return row_hist_scale(d) + 5;
})
.attr("y", function(d, i) {
return i * 12 + 8;
})
.attr("font-family", "sans-serif")
.attr("font-size", "8px")
.attr("fill", "black");
var legend = [0, 2, 4, 6, 8, 10, 12, 14, 16];
svg.append("g")
.attr("transform", "translate(20,60)")
.selectAll("rect")
.data(legend)
.enter()
.append("rect")
.attr("x", function(d, i) {
return i * 12; //Bar width of 12 plus 1 for padding
})
.attr("y", function(d) {
return 20;
})
.attr("width", 10)
.attr("fill",function(d){
return colorPicker(d);
})
.attr("height", 10);
svg.append("g")
.attr("transform", "translate(20,60)")
.selectAll("text")
.data(legend)
.enter()
.append("text")
.text(function(d) {
return d;
})
.attr("x", function(d, i) {
return i * 12 + 5;
})
.attr("y", function(d, i) {
return 40;
})
.attr("font-family", "sans-serif")
.attr("font-size", "8px")
.attr("fill", "black")
.attr("text-anchor", "middle");
svg.append("text").attr("transform", "translate(20,65)")
.text("Legend:")
.attr("font-family", "sans-serif")
.attr("font-size", "12px");
});
<html>
<head>
<script src="http://d3js.org/d3.v3.min.js"></script>
</script>
<link rel="stylesheet" href="d3_heatmap.css" type="text/css"/>
</head>
<body>
<script type="text/javascript" src="d3_heatmap.js"></script>
<script type="text/javascript">
d3.select(self.frameElement).style("height", "300px");
</script>
<div id="table">
</div>
</body>
</html>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment