3. Transposed Visualization with `yame hprint`

yame hprint is a powerful tool for visual inspection of methylation data. It prints the data “horizontally” (transposed), where samples are rows and genomic sites are columns.

It supports two main modes:

Region View: Detailed visualization of a specific genomic interval.
Whole-Genome View: High-level summary of methylation across all chromosomes.

3.1 Basic Usage

yame hprint [options] -R <ref.cr> <in.cx>

-R <ref.cr>: (Required for genomic views) The format 7 reference file used to create the data.
<in.cx>: The input methylation file (supports Format 0, 3, 4, 6, and 7).

3.1.1 Color Output

By default, yame hprint uses ANSI escape codes to provide a rich color visualization.

To disable color (e.g., for piping to text-processing tools), use the -c flag.

3.1.2 Controlling Width and Windowing

If the region or genome being viewed is wider than the terminal (default: 80 columns), yame hprint automatically performs on-the-fly averaging. It calculates the mean beta value for each window of sites to fit the requested width.

-w <int>: Set the maximum number of data columns (default: 80).
-l <int>: Set the width of the sample label column (default: 20).
-t <int>: Ruler tick interval in columns (default: 10).

3.2 View Modes

3.2.1 Region View (`-r`)

Visualize a specific chromosome or sub-region:

# View an entire chromosome
yame hprint -R hg38.cr -r chr16 data.cx

# View a specific coordinate range (1-based)
yame hprint -R hg38.cr -r chr16:10,000,000-10,100,000 data.cx

In region view, the ruler shows coordinates and the number of CpGs included. If windowing is active, it also shows the window size (win=N).

3.2.2 Whole-Genome View

If you provide -R but omit -r, yame hprint shows the entire dataset compressed into the requested width (default 80 columns). This is excellent for identifying large-scale methylation patterns or chromosome-wide changes.

3.3 Granular Decile Mode (`-g`)

By default, yame hprint uses three symbols to represent methylation levels:

H: High beta (> 0.67)
M: Medium beta (0.33 - 0.67)
L: Low beta (< 0.33)
.: Missing / No coverage

For more detail, use the granular mode (-g):

yame hprint -g -R hg38.cr -r chr16 data.cx

In granular mode:

Beta values are mapped to 10 deciles represented by digits 0 to 9.
0: 0.0 - 0.1
5: 0.5 - 0.6
9: 0.9 - 1.0

3.3.1 Blue-White-Red Spectrum

When granular mode and color are active, yame hprint uses a high-quality diverging color map:

Blue (0-4): Progresses from deep blue to light blue (unmethylated).
White (5): Intermediate methylation.
Red (6-9): Progresses from light red to deep red (methylated).

3.4 Technical Implementation

yame hprint is optimized for high performance and low memory footprint:

Two-Pass Scanning: Genomic coordinates are gathered using a two-pass scan of the reference file, avoiding large memory allocations even for very large regions.
On-the-Fly Averaging: Calculations are performed directly from source buffers using offsets.
Zero-Copy: Data is processed in its original memory location without redundant slicing or copying.
Partial Decompression: For Format 3 (.cg), only the required region is decompressed, making it efficient for large files.

3. Transposed Visualization with yame hprint