Visualization

Tranquillyzer provides two visualization tools for inspecting reads and data quality.

Read Annotation Visualization

The visualize command produces color-coded plots of individual reads, showing the model’s predicted annotations for each structural element (adapters, barcodes, UMIs, cDNA, polyA/T). This is useful for spot-checking model predictions and debugging annotation issues.

Visualization runs its own model inference and is independent of annotate-reads. It can be run before, after, or without annotation.

Usage

# Visualize N random reads
tranquillyzer visualize \
    --num-reads 10 \
    --model-name 10x3p_sc_ont_013 \
    --gpu-mem 48 \
    OUTPUT_DIR

# Visualize specific reads by name
tranquillyzer visualize \
    --read-names read_001,read_002,read_003 \
    --model-name 10x3p_sc_ont_013 \
    --gpu-mem 48 \
    OUTPUT_DIR

Either --num-reads or --read-names must be provided.

Options

Option	Default	Description
OUTPUT_DIR	required	Directory with preprocessed reads (writes to plots/)
--output-file	full_read_annots	PDF filename prefix
--model-name	10x3p_sc_ont_013	Model for inference
--num-reads	None	Number of random reads to visualize per Parquet file
--read-names	None	Comma-separated list of specific read names
--gpu-mem	12 GB	GPU memory budget
--target-tokens	1,200,000	Token budget per GPU
--vram-headroom	0.35	VRAM headroom fraction
--threads	2	CPU threads

Output

• plots/{output_file}.pdf:multi-page PDF with one read per page, color-coded by segment type

Read Length Distribution

The readlengthdist command generates a plot showing the distribution of read lengths in your preprocessed data.

tranquillyzer readlengthdist OUTPUT_DIR

Output

• plots/read_length_distribution.png:log10-transformed read lengths on the x-axis, frequencies on the y-axis