# Minimally Reproducible Conda Set Up For Nextflow

**URL:** <https://community.seqera.io/t/minimally-reproducible-conda-set-up-for-nextflow/1279>\
**Category:** Tips & Tricks\
**Tags:** conda\
**Created:** [October 10, 2024, 7:03pm UTC](https://community.seqera.io/t/minimally-reproducible-conda-set-up-for-nextflow/1279 "2024-10-10T19:03:13Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![Sasha\_Dagayev](https://dub1.discourse-cdn.com/flex013/user_avatar/community.seqera.io/sasha_dagayev/32/918_2.png) [@Sasha\_Dagayev](https://community.seqera.io/u/Sasha_Dagayev)\
**Post date:** [October 10, 2024, 7:03pm UTC](https://community.seqera.io/t/minimally-reproducible-conda-set-up-for-nextflow/1279/1 "2024-10-10T19:03:13Z")

</div>

I had a bit of difficulty installing conda via a Dockerfile that would be friendly to Nextflow. Got to these run commands that worked for me:

```auto
# Install Miniconda
RUN curl -O https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh && \
    bash Miniconda3-latest-Linux-x86_64.sh -b -p /opt/conda && \
    rm Miniconda3-latest-Linux-x86_64.sh

# Add conda to PATH
ENV PATH="/opt/conda/bin:${PATH}"

RUN conda init bash
```

---

<div class="post-metadata">

**Author:** ![mahesh.binzerpanchal](https://dub1.discourse-cdn.com/flex013/user_avatar/community.seqera.io/mahesh.binzerpanchal/32/253_2.png) [@mahesh.binzerpanchal](https://community.seqera.io/u/mahesh.binzerpanchal)\
**Post date:** [October 11, 2024, 7:25am UTC](https://community.seqera.io/t/minimally-reproducible-conda-set-up-for-nextflow/1279/2 "2024-10-11T07:25:52Z")

</div>

I would also suggest looking at [Seqera Containers](https://seqera.io/containers/) before trying to make a custom container with Conda ( and not to mention all the business with [conda defaults channel licensing](https://www.datacamp.com/blog/navigating-anaconda-licensing) ).

[Seqera Containers](https://seqera.io/containers/) can build custom images for docker and singularity from conda-forge, bioconda, and pypi using micromamba.

E.g. [Wave container build notification](https://wave.seqera.io/view/builds/f270ba9f06a2df69_1)

If you’re having trouble using the Seqera Containers interface, you can also use the [wave cli](https://github.com/seqeralabs/wave-cli) and build a container that way for example with:

```bash
./wave-1.4.1-linux-x86_64 --freeze --conda-file env.yml

```

The `--freeze` pushes the container to the community registry, and you might be able to use other channels in this way too. The command-line wave interface is also much more powerful and worth exploring.