Step 5. Visualize Data to Assess Gender Deviations

Beyond checking for concordance, visualizing both X and Y chromosome LogRs may provide additional insight into your data. Though plotting this data can be done to a certain degree within HelixTree, it is better to do it in an external program such as Excel.

Figure 1. Joined spreadsheet with
phenotype and X and Y chromosome
LogR means.

To do this you first need to join the reported gender with the X and Y chromosome LogR mean data. The easiest way is to open the joined phenotype and chromosome X LogR mean threshold spreadsheet created in Step 4 and then select >File >Join Spreadsheets on Row Labels, highlight the Chromosome Y LogR mean threshold spreadsheet and click OK. A new spreadsheet will be created with phenotype data and the X and Y chromosome LogR means (Figure 1).

From this joined spreadsheet select >File >Save as Comma-Delimited Text File. Give it a name and click Save.

Open this CSV file in Excel. To plot the data, first sort the spreadsheet by reported gender (in this example, sexC). Next, select all X Mean values for reported females, hold down the Ctrl key and then select all Y Mean values for reported females. With these values selected go to >Insert >Chart.

Figure 2. Completed Step 2 of the Chart
Wizard.

Figure 3. Finished plot of X and Y
chromosome LogRs.

This will bring up Excel's Chart Wizard. Choose a Scatter Plot from Step 1 and click Next. In Step 2, click the Series tab and type ‘Reported Females’ in the empty Name: box. Next, where it says Series click Add to add a second series. Give this the name ‘Reported Males’. For the X Values: select all the X Mean values for reported males. For the Y Values: select all the Y Mean values for reported males. The result of Step 2 in the Chart Wizard should resemble Figure 2.

Click Next to go to Step 3. Here you can enter a title for the chart, and axis names. The X-axis would be X chromosome LogRs and the Y-axis, Y chromosome LogRs. The finished plot should resemble Figure 3.

Again, in this example it is apparent at least one of the samples is an outlier based on X chromosome LogRs and it is possibe a few others are outliers as well.

Additionally, it is important to change the order in which the series are plotted to check whether there are reported females in the reported male cluster or visa-versa. To do this, Right Click on one of the clusters, select Format Data Series and select the Series Order tab. Move series 2 up to see if any blue data points were hidden by the pink cluster. In this case there are not any points hidden by the clusters.