simulate-data CLI

The simulate-data command generates synthetic labeled reads for training. It builds reads according to the library structure defined in seq_orders.yaml, introduces sequencing errors, and outputs the reads and their per-base labels as pickle files in {output_dir}/simulated_data/.

For background on how reads are simulated and why, see Read Structure Simulation and Sequencing Error Modeling.

Command Line Options

CLI Option	Type	Default	What it controls	When to change it
model_name	TEXT	required	Library model key defining segment order and motifs	Set to match your protocol
output_dir	TEXT	required	Output directory (writes to simulated_data/ subdirectory)	—
--training-seq-orders-file	TEXT	utils/seq_orders.yaml	Library definition file specifying segment order and sequences	Only if using a custom file
--num-reads	INT	50,000	Number of reads to simulate (before reverse-complement doubling)	Increase for more training data; decrease for quick tests
--mismatch-rate	FLOAT	0.05	Base substitution probability	Match your sequencing platform’s error profile
--insertion-rate	FLOAT	0.05	Base insertion probability	Match your sequencing platform’s error profile
--deletion-rate	FLOAT	0.06	Base deletion probability	Match your sequencing platform’s error profile
--min-cdna	INT	100	Minimum cDNA length in simulated reads	Adjust to match expected read lengths
--max-cdna	INT	500	Maximum cDNA length in simulated reads	Adjust to match expected read lengths
--polyt-error-rate	FLOAT	0.02	Error rate within polyA/polyT segments	Increase if your data has noisy homopolymers
--max-insertions	INT	1	Maximum insertions allowed after a single base	Rarely needs changing
--threads	INT	2	CPU threads for read simulation	Increase for faster generation on multi-core machines
--rc / --no-rc	FLAG	--rc	Include reverse complements (doubles the training set)	Disable only if your protocol has a fixed known orientation
--transcriptome	TEXT	None	Transcriptome FASTA for realistic cDNA generation	Recommended — improves model accuracy on real data
--invalid-fraction	FLOAT	0.3	Fraction of reads generated as structurally invalid artifacts	Increase if your data has many malformed reads

Example Usage

tranquillyzer simulate-data \
    10x3p_sc_ont \
    training_out \
    --num-reads 50000 \
    --mismatch-rate 0.05 \
    --insertion-rate 0.05 \
    --deletion-rate 0.06 \
    --min-cdna 100 \
    --max-cdna 500 \
    --polyt-error-rate 0.02 \
    --max-insertions 1 \
    --invalid-fraction 0.3 \
    --rc \
    --threads 2 \
    --transcriptome gencode.v44.transcripts.fa