trace_pearsons

Reliability status: stable

Compare multiple chromatin trace tables by computing pairwise distances between barcode combinations and quantifying similarity via Pearson correlation.These can be different fields of view of a single acquisition or different biological replicates.

This tool is useful for analyzing the structural similarity between different chromatin trace datasets, helping to identify patterns and relationships in chromatin organization across multiple samples or conditions.

The script calculates, for each tracefile, the median pairwise distance map. Then it calculates the bin-by-bin Pearson correlation between all tracefiles provided. The result is therefore a matrix of Pearson correlations.

usage: trace_pearsons [-h] (--input INPUT [INPUT ...] | --pipe) [-O OUTPUT]
                      [--vmin VMIN] [--vmax VMAX]
                      [--output_format {png,svg,pdf}] [--verbose]

Named Arguments

--input

Paths to trace files to compare.

--pipe

Read input filenames from stdin (pipe).

Default: False

-O, --output

Output folder name for the correlation matrix plot (filename: trace_correlation_matrix.png)

Default: '.'

--vmin

Minimum value for colormap scaling

Default: -10

--vmax

Maximum value for colormap scaling

Default: 10

--output_format

Possible choices: png, svg, pdf

Output image format. Default = png.

Default: 'png'

--verbose

Increase verbosity

Default: False

Output Files

The script generates:

trace_correlation_matrix.[format]: Pearson correlation matrix plot comparing all input trace tables.

trace_correlation_matrix

trace_correlation_matrix.npy: NPY file containing the Pearson correlation matrix values.

Examples

$ ls *ecsv | trace_pearsons [options]
$ find . -name "*.ecsv" | trace_pearsons [options]

Notes

Input files must be in ECSV format compatible with ChromatinTraceTable
The script identifies unique parts of filenames to create readable labels in the plot
Correlation is calculated based on the spatial distances between barcode pairs