Sequence Collapsing Tool

A Python tool for collapsing identical sequences in FASTA files into unique sequences and restoring them back to their original form.

Features

• Collapses identical DNA/RNA sequences into unique representatives
• Supports field-based grouping for sequence names
• Deterministic output (sequences with the same abundance are alphabetically sorted)
• Tracks abundance and original sequence names
• Reversible process through uncollapsing

Installation

Clone this repository:

git clone https://github.com/MullinsLab/sequence_collapsing.git
cd sequence_collapsing

No additional dependencies required - uses Python standard library only.

Usage

Collapsing Sequences

Basic Usage - Collapse All Sequences

Collapse all identical sequences in a FASTA file:

python collapse_sequences.py input.fasta

Output files:

• input_collapsed.fasta - Collapsed unique sequences
• input_collapsed_name.txt - Mapping file (tab-delimited)

Collapsed sequence naming format: {id}_{uniqueNumber}_{abundance}

Example: sample1_1_5 means the first unique sequence with 5 copies

Advanced Usage - Collapse by Name Fields

Collapse sequences grouped by specific fields in the sequence name:

python collapse_sequences.py input.fasta -i 1,2,3

This groups sequences by fields 1, 2, and 3 of underscore-separated sequence names.

Example: If your sequence names follow the pattern: projectID_sampleID_timepoint_region_other

• -i 1 - Collapse by project only
• -i 1,2 - Collapse by project and sample
• -i 1,2,3 - Collapse by project, sample, and timepoint

Output files:

• input_collapsed_by_field123.fasta
• input_collapsed_by_field123_name.txt

Uncollapsing Sequences

Restore collapsed sequences back to their original form:

python uncollapse_sequences.py collapsed.fasta collapsed_name.txt

Output:

• collapsed_uncollapsed.fasta - Restored original sequences

File Formats

Input FASTA File

Standard FASTA format with sequence names starting with >:

>sequence1_sample_timepoint_region
ATCGATCGATCG
>sequence2_sample_timepoint_region
ATCGATCGATCG
>sequence3_sample_timepoint_region
GCTAGCTAGCTA

Output Collapsed FASTA

Unique sequences with abundance information:

>sample_timepoint_region_1_2
ATCGATCGATCG
>sample_timepoint_region_2_1
GCTAGCTAGCTA

Output Name Mapping File

Tab-delimited file mapping collapsed names to original names:

sample_timepoint_region_1_2    sequence1_sample_timepoint_region,sequence2_sample_timepoint_region
sample_timepoint_region_2_1    sequence3_sample_timepoint_region

Key Features

Deterministic Output

The script ensures reproducible results by sorting sequences with the same abundance alphabetically. Running the same input multiple times will always produce identical output.

Abundance-Based Ordering

Within each group, sequences are ordered by abundance (most common first), with ties broken alphabetically.

Flexible Grouping

The field index option allows flexible grouping strategies based on your sequence naming convention, making it suitable for analyzing sequences across different samples, timepoints, or experimental conditions.

Requirements

• Python 3.x
• Input FASTA files must have .fasta extension

Contact

For any questions, bugs and suggestions, please send email to cohnlabsupport@fredhutch.org and include a few sentences describing, briefly, the nature of your questions and include contact information.