KO | EN

GitLite

Minimalist Open Source Repository Search Engine

gitlite — search

#developer-tools #typescript #claude-code #codex #dsh #dsh-plugin #ai-agent #ai-agents #deepseek-harness #agent-skills #macos #cli

pepper

★ 260 Open GitHub ↗

PEPPER-Margin-DeepVariant

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

## P.E.P.P.E.R.
[![Build Status](https://travis-ci.com/kishwarshafin/pepper.svg?branch=master)](https://travis-ci.com/kishwarshafin/pepper)

`PEPPER` is a genome inference module based on recurrent neural networks that enables long-read variant calling and nanopore assembly polishing in the [PEPPER](https://github.com/kishwarshafin/pepper)-[Margin](https://github.com/UCSC-nanopore-cgl/margin)-[DeepVariant](https://github.com/google/deepvariant) pipeline. This pipeline enables nanopore-based variant calling with [DeepVariant](https://github.com/google/deepvariant).

---

### How to cite
Please cite the following manuscript if you are using `PEPPER-Margin-DeepVariant`:

<details>
<summary><a href="https://www.nature.com/articles/s41592-021-01299-w">Nature Methods: Haplotype-aware variant calling with PEPPER-Margin-DeepVariant enables high accuracy in nanopore long-reads. </a>
</summary>

Authors: Kishwar Shafin, Trevor Pesout, Pi-Chuan Chang, Maria Nattestad, Alexey Kolesnikov, Sidharth Goel, Gunjan Baid, Mikhail Kolmogorov, Jordan M. Eizenga, Karen H. Miga, Paolo Carnevali, Miten Jain, Andrew Carroll & Benedict Paten.

**Free access link to the manuscript:** [https://rdcu.be/cABfv](https://rdcu.be/cABfv)
</details>

---

### Critical care application

In a project led by [Professor Euan Ashley](https://www.euanangusashley.com/), the team demonstrated the ability to identify clinically actionable disease causing variants in a critical care setting with nanopore sequencing and `PEPPER-Margin-DeepVariant`.

Following are the publications that demonstrate the performance of `PEPPER-Margin-DeepVariant` in a clinical setup:

Clinical report:
<details>
<summary><a href="https://www.nejm.org/doi/10.1056/NEJMc2112090">New England Journal of Medicine: Ultrarapid Nanopore Genome Sequencing in a Critical Care Setting </a>
</summary>
Authors: John Gorzynski, Sneha Goenka, Kishwar Shafin, Tanner Jensen, Dianna Fisk, Megan Grove, Elizabeth Spiteri, Trevor Pesout, Jean Monlong, Gunjan Baid, Jonathan Bernstein, Scott Ceresnak, Pi-Chuan Chang, Jeffrey Christle, Henry Chubb, Karen Dalton, Kyla Dunn, Daniel Garalde, Joseph Guillory, Joshua Knowles, Alexey Kolesnikov, Michael Ma, Tia Moscarello, Maria Nattestad, Marco Perez, Maura Ruzhnikov, Mehrzad Samadi, Ankit Setia, Chris Wright, Courtney J Wusthoff, Katherine Xiong, Tong Zhu, Miten Jain, Fritz Sedlazeck, Andrew Carroll, Benedict Paten, Euan Ashley.
</details>

Case report:
<details>
<summary><a href="https://www.ahajournals.org/doi/abs/10.1161/CIRCGEN.121.003591">Circulation: Genomic and Precision Medicine:Ultra-Rapid Nanopore Whole Genome Genetic Diagnosis of Dilated Cardiomyopathy in an Adolescent With Cardiogenic Shock</a>
</summary>
Authors: John Gorzynski, Sneha Goenka, Kishwar Shafin, Tanner Jensen, Dianna Fisk, Megan Grove, Elizabeth Spiteri, Trevor Pesout, Jean Monlong, Jonathan Bernstein, Scott Ceresnak, Pi-Chuan Chang, Jeffrey Christle, Henry Chubb, Kyla Dunn, Daniel Garalde, Joseph Guillory, Maura Ruzhnikov, Chris Wright, Courtney Wusthoff, Katherine Xiong, Seth Hollander, Gerald Berry, Miten Jain, Fritz Sedlazeck, Andrew Carroll, Benedict Paten, Euan Ashley.
</details>

---

### Long read variant calling performance evaluation

Detailed variant calling performance evaluation:
* **Nanopore R9.4.1 Guppy 5.0.6 SUP:**
  * [HG003 whole genome](./docs/performance_evaluation/Oxford_nanopore_r9_whole_genome.md)

Please follow the case-studies documentation for PacBio-HiFi and ONT 10.4 Q20 performance evaluation.

---

### Useful links to documentations
* [Quickstarts to check system configuration.](#quickstarts-small-runs-to-test-system-configuration)
* [Case-studies to reproduce performance.](#case-studies-chromosome-20-runs-for-performance-reproducibility)
* [Description of PEPPER-Margin-DeepVariant method.](./docs/misc/pepper_methods.md)
* How to train PEPPER-DeepVariant:
    * [How to train PEPPER.](./docs/training_pepper_margin_deepvariant/how_to_train_pepper.md)
    * [How to train DeepVariant (using candidates from PEPPER).](./docs/training_pepper_margin_deepvariant/how_to_train_deepvariant.md)
* [How to install PEPPER locally.](./docs/local_install/install_pepper_locally.md)
* [List of parameters and description.](./docs/usage/usage_and_parameters.md)
* [How to tune parameters to balance speed and accuracy.](./docs/usage/speed_and_accuracy.md)

---

### How to run
PEPPER-Margin-DeepVariant can be run using **Docker** or **Singularity**. A simple docker command looks like:
```bash
sudo docker run \
-v "${INPUT_DIR}":"${INPUT_DIR}" \
-v "${OUTPUT_DIR}":"${OUTPUT_DIR}" \
kishwars/pepper_deepvariant:r0.8 \
run_pepper_margin_deepvariant call_variant \
-b "${INPUT_DIR}/${BAM}" \
-f "${INPUT_DIR}/${REF}" \
-o "${OUTPUT_DIR}" \
-t "${THREADS}" \
--ont_r9_guppy5_sup

# --ont_r9_guppy5_sup is preset for ONT R9.4.1 Guppy 5 "Sup" basecaller
# for ONT R10.4 Q20 reads: --ont_r10_q20
# for PacBio-HiFi reads: --hifi
```

### Case studies (chromosome 20 runs for performance reproducibility)

| Pipeline | Docker | Singularity | NVIDIA-docker (GPU) |
|:--------------------------------------------------------------:|:--------------------------------------------------------------:|:-------------------------------------------------------------------------------:|:----------------------------------------------------------------------:|
| Nanopore R9.4.1 variant calling | [Link](./docs/pipeline_docker/ONT_variant_calling.md) | [Link](./docs/pipeline_singularity/ONT_variant_calling_singularity.md) | [Link](./docs/pipeline_docker_gpu/ONT_variant_calling_gpu.md) |
| Nanopore R10.4 Q20 variant calling | [Link](./docs/pipeline_docker/ONT_variant_calling_r10_q20.md) | [Link](./docs/pipeline_singularity/ONT_variant_calling_singularity_r10_q20.md) | [Link](./docs/pipeline_docker_gpu/ONT_variant_calling_r10_q20_gpu.md) |
| PacBio HiFi variant calling | [Link](./docs/pipeline_docker/HiFi_variant_calling.md) | [Link](./docs/pipeline_singularity/HiFi_variant_calling_singularity.md) | [Link](./docs/pipeline_docker_gpu/HiFi_variant_calling_gpu.md) |

### License
[PEPPER license](./LICENSE), [Margin License](https://github.com/UCSC-nanopore-cgl/margin/blob/master/LICENSE.txt) and [DeepVariant License](https://github.com/google/deepvariant/blob/r1.1/LICENSE) extend to the trained models (PEPPER, Margin and DeepVariant) and container environment (Docker and Singularity).

### Acknowledgement
We are thankful to the developers of these packages:
* [htslib & samtools](http://www.htslib.org/)
* [pytorch](https://pytorch.org/)
* [ONNX](https://onnx.ai/)
* [hdf5 python (h5py)](https://www.h5py.org/)

### Authors
[PEPPER](https://github.com/kishwarshafin/pepper)-[Margin](https://github.com/UCSC-nanopore-cgl/margin)-[DeepVariant](https://github.com/google/deepvariant) pipeline is developed in a collaboration between [UC Santa Cruz genomics institute](https://ucscgenomics.soe.ucsc.edu/) and the [Genomics team in Google Health](https://health.google/health-research/genomics/).

### Fun Fact
<img src="https://vignette.wikia.nocookie.net/marveldatabase/images/7/72/Anthony_Stark_%28Earth-616%29_from_Iron_Man_Vol_5_2_002.jpg/revision/latest?cb=20130407031815" alt="Iron-Man" width="240p">

The name "P.E.P.P.E.R." is inspired from an A.I. created by Tony Stark in the  Marvel Comics (Earth-616).

PEPPER is named after Tony Stark's then friend and the CEO of Resilient, Pepper Potts.

pepper

Explore Similar Repositories

Was this content helpful?

Recent Feedback

Download README