Skip to content

Commit

Permalink
updated faq
Browse files Browse the repository at this point in the history
  • Loading branch information
avikdatta committed Dec 12, 2023
1 parent cdad83f commit c7e3ca8
Showing 1 changed file with 49 additions and 37 deletions.
86 changes: 49 additions & 37 deletions analysis_pipelines.md
Original file line number Diff line number Diff line change
Expand Up @@ -5,7 +5,7 @@ title: IGF help pages - Analysis pipelines

<h1>Data Analysis Pipelines</h1>

We now support running lots of community provided pipelines for the projects which are sequenced by us.
We now support running lots of community provided and vendor supplied pipelines for projects which are sequenced by us.


<div class="table-responsive">
Expand All @@ -18,6 +18,11 @@ We now support running lots of community provided pipelines for the projects whi
</tr>
</thead>
<tbody>
<tr>
<td><a href="#faq">FAQ</a></td>
<td></td>
<td></td>
</tr>
<tr>
<td><a href="#snakemake-workflows_rna-seq-star-deseq2">RNA-Seq alignment and DE</a></td>
<td><a href="https://github.com/snakemake-workflows/rna-seq-star-deseq2">snakemake-workflows/rna-seq-star-deseq2</a></td>
Expand Down Expand Up @@ -83,15 +88,54 @@ We now support running lots of community provided pipelines for the projects whi
<td><a href="#nf-core_ampliseq">nf-core/ampliseq</a></td>
<td>Untested</td>
</tr>
</tbody>
</table>
</div>
<p/>
<div>
<h2 id="faq">FAQ</h2>
<table class="table" style="border:hidden;">
<tbody>
<tr>
<td><a href="#faq">FAQ</a></td>
<td></td>
<td></td>
<td style="border:hidden; width:50%"><b>Can we get a custom reference genome for our project?</b></td>
<td style="border:hidden;">Yes, we can build custom reference genome for collaborative projects.</td>
</tr>
<tr>
<td style="border:hidden; width:50%"><b>How do you transfer data?</b></td>
<td style="border:hidden;">We add a new <code>analysis</code> directory to the Globus collection for analysis files. Files are available only for <b>30 days</b>.</td>
</tr>
<tr>
<td style="border:hidden; width:50%"><b>Will you run these pipelines automatically?</b></td>
<td style="border:hidden;">No, we run them for any specific project based on users requests.</td>
</tr>
<tr>
<td style="border:hidden; width:50%"><b>How you run NF-core pipelines on HPC?</b></td>
<td style="border:hidden;">We use a custom configuration file to run NF-core workflows on Imperial College's HPC and track runs via a local installation of <a href="https://github.com/seqeralabs/nf-tower">Nextflow Tower (community edition)</a></td>
</tr>
<tr>
<td style="border:hidden; width:50%"><b>Can we access your Nextflow Tower server?</b></td>
<td style="border:hidden;">We are open for discussions.</td>
</tr>
<tr>
<td style="border:hidden; width:50%"><b>How do you manage multiple pipeline runs on HPC?</b></td>
<td style="border:hidden;">We use <a href="https://airflow.apache.org/">Apache Airflow</a> as orchestration manager and queue multiple pipelines with correct <code>queue</code> and <code>pool</code> information.</td>
</tr>
<tr>
<td style="border:hidden; width:50%"><b>Can you run these pipelines for externally sequenced fastq files?</b></td>
<td style="border:hidden;"><b>NO.</b> We can run them only for the projects which are sequenced by us. Internally, we organise our files following <a href="https://ena-docs.readthedocs.io/en/latest/submit/general-guide/metadata.html">ENA metadata model</a> which helps us to configure these pipelines programmatically. But you can ask for our help if you are trying to run these pipelines on externally sequenced data.</td>
</tr>
<tr>
<td style="border:hidden; width:50%"><b>Can you add XYZ pipeline to this list?</b></td>
<td style="border:hidden;">Yes. We are continuously adding new pipelines to the above list. Feel free to suggest us if your project needs any new analysis pipeline.</td>
</tr>
<tr>
<td style="border:hidden; width:50%"><b>Why some of these pipelines are marked Untested?</b></td>
<td style="border:hidden;">We have checked them only with test data but haven't used them for any real project.</td>
</tr>
</tbody>
</table>
</div>
<p/>
<div align="right"><a href="#table-of-contents">Go to Top</a></div>
<div>
<h2 id="snakemake-workflows_rna-seq-star-deseq2">Snakemake RNA-Seq workflow</h2>
<table class="table" style="border:hidden;">
Expand Down Expand Up @@ -518,35 +562,3 @@ We now support running lots of community provided pipelines for the projects whi
</table>
</div>
<div align="right"><a href="#table-of-contents">Go to Top</a></div>
<div>
<h2 id="faq">FAQ</h2>
<table class="table" style="border:hidden;">
<tbody>
<tr>
<td style="border:hidden; width:50%"><b>Can we get a custom reference genome for our project?</b></td>
<td style="border:hidden;">Yes, we can build custom reference genome for collaborative projects.</td>
</tr>
<tr>
<td style="border:hidden; width:50%"><b>How do you transfer data?</b></td>
<td style="border:hidden;">We add a new <code>analysis</code> directory to the Globus collection for analysis files. Files are available only for <b>30 days</b>.</td>
</tr>
<tr>
<td style="border:hidden; width:50%"><b>Will you run these pipelines automatically?</b></td>
<td style="border:hidden;">No, we run them for any specific project based on users requests.</td>
</tr>
<tr>
<td style="border:hidden; width:50%"><b>How you run NF-core pipelines on HPC?</b></td>
<td style="border:hidden;">We use a custom configuration file to run NF-core workflows on Imperial College's HPC and track runs via a local installation of Nextflow Tower (community edition)</td>
</tr>
<tr>
<td style="border:hidden; width:50%"><b>Can we access your Nextflow Tower server?</b></td>
<td style="border:hidden;">We are open for discussions.</td>
</tr>
<tr>
<td style="border:hidden; width:50%"><b>How do you manage multiple pipeline runs on HPC?</b></td>
<td style="border:hidden;">We use Apache Airflow as orchestration manager and queue multiple pipelines with correct <code>queue</code> and <code>pool</code> information.</td>
</tr>
</tbody>
</table>
</div>
<div align="right"><a href="#table-of-contents">Go to Top</a></div>

0 comments on commit c7e3ca8

Please sign in to comment.