Systematic Analysis of DNA Microarray Data: Ordering and Interpreting Patterns of Gene Expression

(Downloading may take up to 30 seconds. If the slide opens in your browser, select File -> Save As to save it.)

Click on image to view larger version.

Figure 5.
Figure 5.

Recoding continuous data by binning. The far left column shows the initial raw data (raw). The column to the right of this shows the raw data rounded to the nearest whole integer (rounded). Recoding these attributes using BINNING 1 uses two bins that are .4 units wide, with a floating (missing) region of values between them that is .2 units wide. Recoding using BINNING 2 uses two bins that are .2 units wide with a large floating region between that is .6 units. Recoding using BINNING 3 has three bins that are .3 units wide with gaps between the bins that are .1 units wide. There are any number of arbitrary ways that bins can be erected. Note that the floating values between the bins are scored as “?”. This method of scoring in phylogenetic analysis implies that the data are missing and will have no impact on the outcome of the parsimony tree.

This Article

  1. Genome Res. 11: 1149-1155

Preprint Server