ratih 1.0.0a1

RATIH: Lunar Transcriptome Assembly - Fast, modern replacement for Trinity

🌙 RATIH - Lunar Transcriptome Assembly

[] [] []

RATIH (Robust Assembler for Transcriptomes with Integrated Hashing) is a modern, high-performance de novo transcriptome assembler.

FYI: RATIH named after the Javanese-Balinese Hindu goddess of the moon, symbol of love and beauty.

🌟 Features

• 🚀 10-100x faster than Trinity
• 📦 Zero R dependencies - pure Python statistical stack
• 🔧 One-line install: pip install ratih
• 🎯 Drop-in replacement for Trinity + Trinotate
• 📊 Interactive web dashboard
• 🧬 Complete annotation pipeline (BLAST, PFAM, SignalP, TmHMM, EggNOG)
• 📈 Differential expression (edgeR, DESeq2, limma)
• 🎨 Lunar phases: Crescent → Gibbous → FullMoon

📦 Installation

# Basic install
pip install ratih

# With full features
pip install ratih[full]

# With web dashboard
pip install ratih[web]

# Development install
git clone https://gitlab.com/biomikalab/ratih
cd ratih
make develop

🚀 Quick Start

Command Line

# Full pipeline
ratih assemble -1 reads_R1.fastq -2 reads_R2.fastq -o results/

# Phase by phase
ratih crescent assemble -r reads.fasta
ratih gibbous assemble -c crescent_out/contigs.fa -r reads.fasta
ratih fullmoon assemble -g gibbous_out/graph.txt -r gibbous_out/reads.txt

# Annotation
ratih annotate -t transcripts.fa -b blast.out -o annotation/

# Differential expression
ratih analyze de -c counts.csv -s samples.csv -m edgeR

# Launch web dashboard
ratih web --port 8080

Python API

from ratih import Pipeline, LunarAssembler, TrinotateAnnotator

# Complete pipeline
pipeline = Pipeline(
    left_reads="sample_R1.fastq",
    right_reads="sample_R2.fastq",
    output_dir="my_assembly"
)
results = pipeline.run()

print(f"Assembled {len(results.transcripts)} transcripts")
print(f"N50: {results.stats.n50}")

# Or phase by phase
assembler = LunarAssembler(kmer_size=25, cpu=16)
transcripts = assembler.assemble("reads_R1.fastq", "reads_R2.fastq")

# Annotate
annotator = TrinotateAnnotator(transcripts_fasta="assembly.fa")
annotator.run()

📊 Performance

Metric	Trinity	RATIH	Improvement
Assembly time (100M reads)	8-12 hours	1-2 hours	6-8x faster
Memory usage	32-64 GB	4-8 GB	4-8x less
Installation	Complex (Docker)	pip install	Simpler!
R dependencies	Required	None	Zero

🌙 Assembly by Lunar Phases

Following example of Trinity (but of course, improved), the assembly follows the lunar cycle:

🌙 Crescent - Initial contig assembly (replaces Inchworm)

🌖 Gibbous - De Bruijn graph construction (replaces Chrysalis)

🌕 FullMoon - Final transcript extraction (replaces Butterfly)

📚 Documentation

Full documentation is available at ratih.readthedocs.io

🤝 Contributing

Contributions are welcome! Please see CONTRIBUTING.md

📄 License

RATIH is dual-licensed under:

• BSD 3-Clause License - Permissive, allows proprietary use
• GNU Affero General Public License v3.0 (AGPL-3.0) - Copyleft, requires source disclosure for network services

You may choose either license at your option.

When to use which license?

Use Case	Recommended License
Proprietary software integration	BSD-3-Clause
Academic research	Either
Network service/API	AGPL-3.0
Open source projec	Either
Commercial SaaS	AGPL-3.0

See LICENSE-BSD and LICENSE-AGPL for details.

SPDX-License-Identifier: BSD-3-Clause OR AGPL-3.0-only

🌟 Citation If you use RATIH in your research, please cite:

Nashrulloh, M.M., RATIH contributors. (2026). RATIH: Lunar Transcriptome Assembly - Fast, modern replacement for Trinity (Version 1.0.0) [Computer software]. https://gitlab.com/biomikalab/ratih