/// <summary>
/// Generate a histogram of GC content over all sequences.
/// </summary>
/// <param name="filename">Filename of the output image file</param>
/// <param name="height">Height of the plot in pixels</param>
/// <param name="width">Width of the plot in pixels</param>
public void PlotGCContentBySequence(string filename, int height = ShoHelper.HeightDefault, int width = ShoHelper.WidthDefault)
{
if (this.Analyzer.GCContentBySequenceArray == null)
{
throw new ArgumentNullException("this.analyzer.SymbolCountByPositionTable");
}
int binSize = 5;
var bins = ShoHelper.CreateBins(0, 100, binSize);
var hist = new Histogram(this.Analyzer.GCContentBySequenceArray, bins);
ShoChart f = ShoHelper.CreateBasicShoChart(
false,
height,
width,
"Number of reads",
"GC content (%; bin size = " + binSize + ")",
"GC content by sequence " + titleBanner
);
var xLabelPositions = ShoHelper.CreateBins(1, 21);
f.SetXLabels(xLabelPositions, bins);
f.Bar(bins, hist.Count);
//f.DundasChart.Series[0].ChartType = SeriesChartType.Spline;
f.SaveImage(filename);
}
/// <summary>
/// Generate a plot of quality score by position
/// </summary>
/// <param name="filename">Filename of the output image file</param>
/// <param name="height">Height of the plot in pixels</param>
/// <param name="width">Width of the plot in pixels</param>
public void PlotQualityScoreCountByPosition(string filename, int height = ShoHelper.HeightDefault, int width = ShoHelper.WidthDefault)
{
if (!this.Analyzer.HasRunContentByPosition)
{
throw new ArgumentException("Unable to plot. Need to process quality score data first.");
}
ShoChart f = ShoHelper.CreateBasicShoChart(
false,
height,
width,
"Phred quality score (" + this.Analyzer.FormatType.ToString() + ")",
"Read position (bp)",
""
);
// Set x- and y-axis ranges
//f.SetXRange(1, this.Analyzer.ReadLengthMax);
//f.SetYRange(Convert.ToDouble(QualitativeSequence.GetMinQualScore(this.Analyzer.FormatType)),
// Convert.ToDouble(QualitativeSequence.GetMaxQualScore(this.Analyzer.FormatType)));
var xLabels = ShoHelper.CreateBins(1, (int)this.Analyzer.ReadLengthMax);
f.SetXLabels(xLabels, null);
// Create a BoxPlot by reading data previously saved to file via FileStream.
BoxPlot bp = new BoxPlot(f, (int)this.Analyzer.ReadLengthMax, this.Analyzer.Count, this.worker, this.workerArgs);
// Update chart title
f.Title = string.Format("Base quality score by position {0}", MakeTitleBanner(Path.GetFileName(this.Analyzer.FileName), bp.NumberOfProcessedReads));
// Build boxplot
bp.BuildBoxPlot(xLabels, this.Analyzer.MemStream);
f.SaveImage(filename);
}
/// <summary>
/// Generate a plot of symbol count by position using Sho libraries.
/// Also include GC count, if applicable.
/// </summary>
/// <param name="filename">Filename of the output image file</param>
/// <param name="height">Height of the plot in pixels</param>
/// <param name="width">Width of the plot in pixels</param>
public void PlotSymbolCountByPosition(string filename, int height = ShoHelper.HeightDefault, int width = ShoHelper.WidthDefault)
{
if (this.Analyzer.SymbolCountByPositionTable == null)
{
throw new ArgumentNullException("SymbolCountByPositionTable");
}
ShoChart f = ShoHelper.CreateBasicShoChart(
true,
height,
width,
"Percentage (%)",
"Read position (bp)",
"Base content by position " + titleBanner
);
// Set additional ShoChart parameters
f.SetYRange(0, 100);
f.SetXRange(0, this.Analyzer.ReadLengthMax + 1);
var xLabels = ShoHelper.CreateBins(1, (int)this.Analyzer.ReadLengthMax);
f.SetXLabels(xLabels, null);
// Calculate the total number of reads by position
int[] totalCountByPosition = this.Analyzer.GetSumByPosition();
// Plot symbol content for each base (i.e. A, C, G, T)
int seriesCount = 0;
foreach (KeyValuePair <byte, int[]> pair in this.Analyzer.SymbolCountByPositionTable)
{
// convert counts to percentages
double[] percentages = new double[this.Analyzer.ReadLengthMax];
for (int i = 0; i < this.Analyzer.ReadLengthMax; i++)
{
int countAtCurrentPosition = pair.Value.Sum();
percentages[i] = pair.Value[i] * 100 / totalCountByPosition[i];
}
// plot the series
f.Plot(xLabels, percentages, "-" + seriesColors[seriesCount]);
f.DundasChart.Series[seriesCount].ChartType = SeriesChartType.StackedArea100;
// set the series name (for legend display)
f.SeriesNames[seriesCount] = System.Text.Encoding.UTF8.GetString(new byte[] { pair.Key });
seriesCount++;
}
//f.DundasChart.Series[0].ChartType = System.Windows.Forms.DataVisualization.Charting.SeriesChartType.StackedColumn100;
//f.DundasChart.AlignDataPointsByAxisLabel();
// If applicable, plot GC content by position as well
if (this.Analyzer.Alphabet == Alphabets.DNA ||
this.Analyzer.Alphabet == Alphabets.AmbiguousDNA ||
this.Analyzer.Alphabet == Alphabets.RNA ||
this.Analyzer.Alphabet == Alphabets.AmbiguousRNA)
{
var gcContentByPosition = this.Analyzer.GCContentByPositionArray;
f.Plot(xLabels, gcContentByPosition, ":ok");
f.SeriesNames[seriesCount] = "GC%";
f.DundasChart.Series[seriesCount].ChartType = SeriesChartType.Spline;
}
// save plot to image file
f.SaveImage(filename);
}
