This work is NOT published as-of yet, and is in progress. Please stay tuned for the published version of our software.

Shield:

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

TAUSO

Feature extraction and analysis utilities for antisense oligonucleotides (ASO) design, built for the TAU-Israel 2025 iGEM project.

Features

• MOE (20-mer) and LNA (16-mer) candidate ranking pipeline with feature breakdown and off-target analysis.
• Features include GC content, hybridization, RNA accessibility, folding, and toxicity heuristics.
• Tool to download genomic sequences, annotations, and index them with bowtie.

Getting Started

→ New users: See Quick Start Guide for step-by-step instructions.

Installation

# One command - installs everything
mamba env create -f environment-dev.yml
mamba activate tauso

→ Complete guide: Installation Guide

Data Setup

Features are computed against a locally built genome, off-target index, and omics datasets — on the order of 10 GB, most of it the genome and the bowtie off-target index. The build is dominated by downloading the genome and indexing it; budget the better part of an hour.

tauso setup-all              # genome, bowtie off-target index, omics, raccess
tauso build-cell-context     # per-cell expression + CAI weights + tGCN

→ Complete guide: Data Setup Guide

Usage

The pipeline ranks MOE (20-mer) and LNA (16-mer) candidates against a target, with a per-feature breakdown and off-target analysis. Feature computation is the bottleneck in this version — a full target can take tens of minutes — and performance work is ongoing.

The public API is not yet settled. Feature implementations live under tauso.features and can be explored directly. Pass a custom_sequence to analyze an exogenous or mutated target. A documented, stable entry point will accompany the article.

Collaborations

If you want to work with us, or access the future paper, please feel free to contact us at igem.tau.official@gmail.com