Venn diagrams
<head>
<script type="text/javascript" src="http://www.google.com/jsapi"></script> <script type="text/javascript"> google.load("visualization", "1", {packages:["barchart"]}); google.setOnLoadCallback(drawChart); function drawChart() { var data = new google.visualization.DataTable(); data.addColumn('string', 'Annotation'); data.addColumn('number', 'All three call the same stop codon index but rast has different start codon');
data.addColumn('number', 'All three call Different stop codon index');
data.addRows(3);
data.setValue(0, 0, 'JGI');
data.setValue(0, 1,650);
data.setValue(0, 2, 290);
data.setValue(1, 0, 'Rast');
data.setValue(1, 1,682);
data.setValue(1, 2,472);
data.setValue(2, 0, 'Manatee');
data.setValue(2, 1,655);
data.setValue(2, 2,242);
var chart = new google.visualization.BarChart(document.getElementById('chart_div'));
chart.draw(data, {width: 800, height: 300, is3D: true,titleX:"Average Gene Size (bp)", title: 'Average Gene Size Comparison'});
}
</script>
</head>
<body>
</body>
Analysis
The major trend that can be noted by comparing these 2 Venn diagrams involves the Rast annotation. In the Figure 1, JGI and Manatee overlapped with each other (985+1471=2456) significantly more than 3 annotations overlapped collectively (1471). This means that the gene calls in Rast significantly differed from the gene calls made by JGI and Manatee. However, in Figure 2 this trend disappears. This is likely due to the use of more alternative start codons by Rast. While the Rast's stop codons calls were usually the same as JGI and Manatee, the start codons that Rast called frequently differed from the other two annotations. This often resulted in Rast predicting longer genes, as can be view in this histogram.
