Hitmaps

Hitmaps are a type of data visualization that displays two-dimensional arrays where each cell's color encodes the magnitude of hits, matches, or events associated with the corresponding row and column categories. The coordinates on the axes typically represent discrete categories, such as sequence identifiers, genomic positions, time points, or query-reference pairs. The value in each cell can be a count, a frequency, a p-value, or a normalized score; common practice uses either linear or log-scaled color scales and often a diverging palette when the data center around a meaningful baseline.

Construction involves building a matrix by counting occurrences of hits; apply normalization (per-row, per-column, or global); optionally aggregate data into bins; choose a color scale; optionally add annotations or a color legend. The result is a compact visualization that highlights densities, clusters, or patterns across the two dimensions.

Applications are diverse. In genomics, hitmaps visualize where sequence reads or motifs occur along a chromosome or across multiple samples. In text analysis or information retrieval, they can show the density of keyword hits across documents or corpora. In network analysis, hitmaps may represent co-occurrence counts between items or events over time or categories.

Advantages include an intuitive overview and the ability to reveal cluster structure and relative emphasis across many categories. Limitations involve sensitivity to binning and scaling choices, potential loss of detail in dense data, and reliance on color perception. Hitmaps are most effective when the goal is to compare relative magnitudes across many category pairs at a glance.