# From peaks to genes

### Overview

question Questions
• How to use Galaxy?

• How to get from peak regions to a list of gene names?

objectives Objectives
• Familiarize yourself with the basics of Galaxy

• Learn how to obtain data from external sources

• Learn how to run tools

• Learn how histories work

• Learn how to create a workflow

• Learn how to share your work

time Time estimation: 3 hours

level Level: Introductory level level level

Supporting Materials

last_modification Last modification: Dec 1, 2020

# Introduction

We stumbled upon a paper (Li et al. 2012) called “The histone acetyltransferase MOF is a key regulator of the embryonic stem cell core transcriptional network”. The paper contains the analysis of possible target genes of an interesting protein called Mof. The targets were obtained by ChIP-seq in mice and the raw data is available through GEO. However, the list of genes is neither in the supplement of the paper, nor part of the GEO submission. The closest thing we could find is a file in GEO containing a list of the regions where the signal is significantly enriched (so called peaks):

 1 3660676 3661050 375 210 62.0876 -2.00329 1 3661326 3661500 175 102 28.2951 -0.695557 1 3661976 3662325 350 275 48.3063 -1.29391 1 3984926 3985075 150 93 34.188 -0.816992 1 4424801 4424900 100 70 26.8023 -0.66282

Table 1 Subsample of the available file

The goal of this exercise is to turn this list of genomic regions into a list of possible target genes.

### comment Note: results may vary

Your results may be slightly different from the ones presented in this tutorial due to differing versions of tools, reference data, external databases, or because of stochastic processes in the algorithms.

### Agenda

In this tutorial, we will deal with:

1. Pretreatments
2. Part 1: Naive approach
3. Part 2: More sophisticated approach

# Pretreatments

### hands_on Hands-on: Open Galaxy

1. Browse to a Galaxy instance: the one recommended by your instructor or one in the list Galaxy instance on the head of this page

The Galaxy interface consist of three main parts. The available tools are listed on the left, your analysis history is recorded on the right, and the central panel will show the tools and datasets.

### hands_on Hands-on: Create history

1. Make sure you have an empty analysis history.

### tip Tip: Creating a new history

Click the new-history icon at the top of the history panel

If the new-history is missing:

1. Click on the galaxy-gear icon (History options) on the top of the history panel
2. Select the option Create New from the menu
2. Rename your history to make it easy to recognize

### tip Rename a history

• Click on the title of the history (by default the title is Unnamed history)

• Type Galaxy Introduction as the name
• Press Enter

1. Download the list of peak regions (the file GSE37268_mof3.out.hpeak.txt.gz) from GEO to your computer
2. Click on the upload button in the upper left ot the interface

3. Press Choose local file and search for your file on your computer
4. Select interval as Type
5. Press Start
6. Press Close
7. Wait for the upload to finish. Galaxy will automatically unpack the file.

8. After this you will see your first history item in Galaxy’s right pane. It will go through the gray (preparing/queued) and yellow (running) states to become green (success):

Directly uploading files is not the only way to get data into Galaxy

• Open the Galaxy Upload Manager (galaxy-upload on the top-right of the tool panel)

• Select Paste/Fetch Data
• Paste the link into the text field

• Change Type from “Auto-detect” to interval

• Press Start

• Close the window

By default, Galaxy uses the URL as the name, so rename the files with a more useful name.

### tip Tip: Importing data to Galaxy

There are more options for advanced users.

### comment Interval file format

Interval format is a Galaxy format for representing genomic intervals. It is tab-separated, but has the added requirement that three of the columns must be:

• chromosome ID
• start position (0-based)
• end position (end-exclusive)

An optional strand column can also be specified, and an initial header row can be used to label the columns, which do not have to be in any special order. Unlike BED format (see below) arbitrary additional columns can also be present.

You can find more information about formats that can be used in Galaxy at the Galaxy Data Formats page.

### hands_on Hands-on: Inspect and edit attributes of a file

1. Click on the file in the history panel

Some meta-information (e.g. format, reference database) about the file and the header of the file are then displayed, along with the number of lines in the file (48,647):

2. Click on the galaxy-eye (eye) icon (View data) in your dataset in the history

The content of the file is displayed in the central panel

3. Click on the galaxy-pencil (pencil) icon (Edit attributes) in your dataset in the history

A form to edit dataset attributes is displayed in the central panel

4. Search for mm9 in Database/Build attribute and select Mouse July 2007 (NCBI37/mm9) (the paper tells us the peaks are from mm9)
5. Click on Save on the top
6. Add a tag called #peaks to the dataset to make it easier to track in the history

### tip Tip: Adding a tag

• Click on the dataset
• Click on galaxy-tags Edit dataset tags
• Add a tag starting with #

Tags starting with # will be automatically propagated to the outputs of tools using this dataset.

• Check that the tag is appearing below the dataset name

The dataset should now look like below in the history

In order to find the related genes to these peak regions, we also need a list of genes in mice, which we can obtain from UCSC.

### hands_on Hands-on: Data upload from UCSC

1. Search for UCSC Main in the tool search bar (top left)

2. Click on UCSC Main tool

You will be taken to the UCSC table browser, which looks something like this:

3. Set the following options:
• “clade”: Mammal
• “genome”: Mouse
• “assembly”: July 2007 (NCBI37/mm9)
• “group”: Genes and Gene Predictions
• “track”: RefSeq Genes
• “table”: refGene
• “region”: genome
• “output format”: BED - browser extensible data
• “Send output to”: Galaxy (only)
4. Click on the get output button

You will see the next screen:

5. Make sure that “Create one BED record per” is set to Whole Gene
6. Click on the Send Query to Galaxy button
7. Wait for the upload to finish
8. Rename our dataset to something more recognizable like Genes

### tip Tip: Renaming a dataset

• Click on the galaxy-pencil pencil icon for the dataset to edit its attributes
• In the central panel, change the Name field to Genes
• Click the Save button
9. Add a tag called #genes to the dataset to make it easier to track in the history

### commentBED file format

The BED - Browser Extensible Data format provides a flexible way to encode gene regions. BED lines have three required fields:

• chromosome ID
• start position (0-based)
• end position (end-exclusive)

There can be up to and nine additional optional fields, but the number of fields per line must be consistent throughout any single set of data.

Now we have collected all the data we need to start our analysis.

# Part 1: Naive approach

We will first use a “naive” approach to try to identify the genes that the peak regions are associated with. We will identify genes that overlap at least 1bp with the peak regions.

## File preparation

Let’s have a look at our files to see what we actually have here.

### hands_on Hands-on: View file content

1. Click on the galaxy-eye (eye) icon (View data) of the peak file to view the content of it

It should look like this:

2. View the content of the regions of the genes from UCSC

### question Questions

While the file from UCSC has labels for the columns, the peak file does not. Can you guess what the columns stand for?

### solution Solution

This peak file is not in any standard format and just by looking at it, we cannot find out what the numbers in the different columns mean. In the paper the authors mention that they used the peak caller HPeak.

By looking at the HPeak manual we can find out that the columns contain the following information:

• chromosome name by number
• start coordinate
• end coordinate
• length
• location within the peak that has the highest hypothetical DNA fragment coverage (summit)
• not relevant
• not relevant

In order to compare the two files, we have to make sure that the chromosome names follow the same format. As we can see, the peak file lacks chr before any chromosome number. But what happens with chromosome 20 and 21? Will it be X and Y instead? Let’s check:

### hands_on Hands-on: View end of file

1. Search for Select last tool tool and run Select last lines from a dataset (tail) with the following settings:
• “Text file”: our peak file GSE37268_mof3.out.hpeak.txt.gz
• “Operation”: Keep last lines
• “Number of lines”: Choose a value, e.g. 100
2. Click Execute
3. Wait for the job to finish
4. Inspect the file through the galaxy-eye (eye) icon (View data)

### question Questions

1. How are the chromosomes named?
2. How are the chromosomes X and Y named?

### solution Solution

1. The chromosomes are just given by their number. In the gene file from UCSC, they started with chr
2. The chromosomes X and Y are named 20 and 21

In order to convert the chromosome names we have therefore two things to do:

1. add chr
2. change 20 and 21 to X and Y

### hands_on Hands-on: Adjust chromosome names

1. Replace Text Tool: toolshed.g2.bx.psu.edu/repos/bgruening/text_processing/tp_replace_in_column/1.1.3 in a specific column with the following settings:
• “File to process”: our peak file GSE37268_mof3.out.hpeak.txt.gz
• “in column”: 1
• “Find pattern”: [0-9]+

This will look for numerical digits

• “Replace with”: chr&

& is a placeholder for the find result of the pattern search

2. Rename your output file chr prefix added.

3. Replace Text Tool: toolshed.g2.bx.psu.edu/repos/bgruening/text_processing/tp_replace_in_column/1.1.3 : Let’s rerun the tool with
• “File to process”: the output from the last run, chr prefix added
• “in column”: 1
• “Find pattern”: chr20
• “Replace with”: chrX

### tip Tip: Rerunning a tool

• Expand the dataset information
• Press the galaxy-refresh icon (Run this job again)
4. Rename your output file chrX fixed

5. Replace Text Tool: toolshed.g2.bx.psu.edu/repos/bgruening/text_processing/tp_replace_in_column/1.1.3 : Rerun this tool to do the same for chromosome Y
• “File to process”: chrX fixed, the output from the last run
• “in column”: 1
• “Find pattern”: chr21
• “Replace with”: chrY
6. Inspect the latest file through the galaxy-eye (eye) icon. Have we been successful?

We have quite a few files now and need to take care to select the correct ones at each step.

### question Questions

How many regions are in our output file? You can click the name of the output to expand it and see the number.

### solution Solution

It should be equal to the number of regions in your first file, GSE37268_mof3.out.hpeak.txt.gz: 48,647 If yours says 100 regions, then you have run it on the Tail file and need to re-run the steps.

7. Rename the file to something more recognizable, e.g. Peak regions

## Analysis

Our goal is to compare the 2 region files (the genes file and the peak file from the publication) to know which peaks are related to which genes. If you only want to know which peaks are located inside genes (within the gene body) you can skip the next step. Otherwise, it might be reasonable to include the promoter region of the genes into the comparison, e.g. because you want to include transcriptions factors in ChIP-seq experiments. There is no strict definition for promoter region but 2kb upstream of the TSS (start of region) is commonly used. We’ll use the Get Flanks tool to get regions 2kb bases upstream of the start of the gene to 10kb bases downstream of the start (12kb in length). To do this we tell the Get Flanks tool we want regions upstream of the start, with an offset of 10kb, that are 12kb in length, as shown in the diagram below.

### hands_on Hands-on: Add promoter region to gene records

1. Get Flanks Tool: toolshed.g2.bx.psu.edu/repos/devteam/get_flanks/get_flanks1/1.0.0 returns flanking region/s for every gene, with the following settings:
• “Select data”: Genes file from UCSC
• “Region”: Around Start
• “Location of the flanking region/s”: Upstream
• “Offset”: 10000
• “Length of the flanking region(s)”: 12000

This tool returns flanking regions for every gene

2. Compare the rows of the resulting BED file with the input to find out how the start and end positions changed

### tip Tip: Inspecting several files using the scratchbook

• Click Enable/Disable Scratchbook on the top panel

• Click on the galaxy-eye (eye) icon of the files to inspect
• Click on Show/Hide Scratchbook

3. Rename your dataset to reflect your findings (Promoter regions)

The output is regions that start from 2kb upstream of the TSS and include 10kb downstream. For input regions on the positive strand e.g. chr1 134212701 134230065 this gives chr1 134210701 134222701. For regions on the negative strand e.g. chr1 8349819 9289958 this gives chr1 9279958 9291958.

You might have noticed that the UCSC file is in BED format and has a database associated to it. That’s what we want for our peak file as well. The Intersect tool we will use can automatically convert interval files to BED format but we’ll convert our interval file explicitly here to show how this can be achieved with Galaxy.

### hands_on Hands-on: Change format and database

1. Click on the galaxy-pencil (pencil) icon in the history entry of our peak region file
2. Switch to the Convert tab
3. Select Convert Genomic Intervals to BED
4. Press Convert datatype
5. Check that the “Database/Build” is mm9 (the database build for mice used in the paper)
6. Again rename the file to something more recognizable, e.g. Peak regions BED

It’s time to find the overlapping intervals (finally!). To do that, we want to extract the genes which overlap/intersect with our peaks.

### hands_on Hands-on: Find Overlaps

1. Intersect Tool: toolshed.g2.bx.psu.edu/repos/devteam/intersect/gops_intersect_1/1.0.0 the intervals of two datasets, with the following settings:
• “Return”: Overlapping Intervals
• “of”: the UCSC file with promoter regions (Promoter regions)
• “that intersect”: our peak region file from Replace (Peak regions BED)
• “for at least”: 1

The order of the inputs is important! We want to end up with a list of genes, so the corresponding dataset with the gene information needs to be the first input (Promoter regions).

We now have the list of genes (column 4) overlapping with the peak regions, similar to shown above.

To get a better overview of the genes we obtained, we want to look at their distribution across the different chromosomes. We will group the table by chromosome and count the number of genes with peaks on each chromosome

### hands_on Hands-on: Count genes on different chromosomes

1. Group Tool: Grouping1 data by a column and perform aggregate operation on other columns with the following settings:
• “Select data” to the result of the intersection
• “Group by column”:Column 1
• Press Insert Operation and choose:
• “Type”: Count
• “On column”: Column 1
• “Round result to nearest integer?”: No

### question Questions

Which chromosome contained the highest number of target genes?

### solution Solution

The result varies with different settings, for example, the annotation may change due to updates at UCSC. If you followed step by step, with the same annotation, it should be chromosome 11 with 1992 genes. Note that for reproducibility, you should keep all input data used because Galaxy can store all parameters but inputs may change e.g. the annotation from UCSC.

## Visualization

Since we have some nice data, let’s draw a barchart out of it!

### hands_on Hands-on: Draw barchart

1. Click on galaxy-barchart (visualize) icon on the output from the Group tool
2. Select Bar diagram
3. Choose a title at Provide a title, e.g. Gene counts per chromosome
4. Switch to the galaxy-chart-select-data Select data tab and play around with the settings
5. When you are happy, click the galaxy-save Save visualization in the top right of the main panel

This will store it to your saved visualisations where you can later view, download, or share it with others.

## Extracting workflow

When you look carefully at your history, you can see that it contains all steps of our analysis, from the beginning to the end. By building this history we have actually built a complete record of our analysis with Galaxy preserving all parameter settings applied at every step. Wouldn’t it be nice to just convert this history into a workflow that we’ll be able to execute again and again?

Galaxy makes this very simple with the Extract workflow option. This means that any time you want to build a workflow, you can just perform it manually once, and then convert it to a workflow, so that next time it will be a lot less work to do the same analysis. It also allows you to easily share or publish your analysis.

### hands_on Hands-on: Extract workflow

If you had any failed jobs (red), please remove those datasets from your history by clicking on the galaxy-cross (cross) button. This will make the creation of a workflow easier.

2. Go to the history History options (galaxy-gear (gear) icon)
3. Select the Extract Workflow option.

The central panel will change and you will be able to choose which steps to include/exclude and how to name the newly created workflow.

4. Uncheck any steps that shouldn’t be included in the workflow

Since we did some steps which where specific to our custom peak file, we might want to exclude:

• Select last tool
• all Replace Text tool steps
• Convert Genomic Intervals to BED
• Get flanks tool
5. Rename the workflow to something descriptive, e.g. From peaks to genes

6. Click on the Create Workflow button near the top.

You will get a message that the workflow was created. But where did it go?

7. Click on Workflow in the top menu of Galaxy

Here you have a list of all your workflows

8. Select the newly generated workflow and click on Edit

You should see something similar to this:

### comment The workflow editor

We can examine the workflow in Galaxy’s workflow editor. Here you can view/change the parameter settings of each step, add and remove tools, and connect an output from one tool to the input of another, all in an easy and graphical manner. You can also use this editor to build workflows from scratch.

Although we have our two inputs in the workflow they are missing their connection to the first tool (Intersect tool), because we didn’t carry over some of the intermediate steps.

9. Connect each input dataset to the Intersect tool tool by dragging the arrow pointing outwards on the right of its box (which denotes an output) to an arrow on the left of the Intersect box pointing inwards (which denotes an input)
10. Rename the input datasets to Reference regions and Peak regions
11. Click on the galaxy-gear (gear) icon at the top right
12. Press Auto Re-layout to clean up our view
13. Click on the galaxy-gear (gear) icon at the top right
14. Press Save to save your changes

### tip Tip: Hiding intermediate steps

When a workflow is executed, the user is usually primarily interested in the final product and not in all intermediate steps. By default all the outputs of a workflow will be shown, but we can explicitly tell Galaxy which output to show and which to hide for a given workflow. This behaviour is controlled by the little asterisk next to every output dataset:

If you click on this asterisk for any of the output datasets, then only files with an asterisk will be shown, and all outputs without an asterisk will be hidden (Note that clicking all outputs has the same effect as clicking none of the outputs, in both cases all the datasets will be shown).

Now it’s time to reuse our workflow for a more sophisticated approach.

# Part 2: More sophisticated approach

In part 1 we used an overlap definition of 1 bp (default setting) to identify genes associated with the peak regions. However, the peaks could be broad, so instead, in order to get a more meaningful definition, we could identify the genes that overlap where most of the reads are concentrated, the peak summit. We will use the information on the position of the peak summit contained in the original peak file and check for overlap of the summits with genes.

## Preparation

We again need our peak file, but we’d like to work in a clean history. Instead of uploading it twice, we can copy it to a new history.

### hands_on Hands-on: Copy history items

1. Create a new history and give it a new name like Galaxy Introduction Part 2

### tip Tip: Creating a new history

Click the new-history icon at the top of the history panel

If the new-history is missing:

1. Click on the galaxy-gear icon (History options) on the top of the history panel
2. Select the option Create New from the menu
2. Click on the View all histories (galaxy-columns icon) at the top right of your history

You should see both of your histories side-by-side now

3. Drag and drop the edited peak file (Peak regions, after the replace steps), which contains the summit information, to your new history.
4. Click on Analyze Data in the top menu bar to go back to your analysis window

## Create peak summit file

We need to generate a new BED file from the original peak file that contains the positions of the peak summits. The start of the summit is the start of the peak (column 2) plus the location within the peak that has the highest hypothetical DNA fragment coverage (column 5). As the end of the peak region, we will simply define start + 1.

### hands_on Hands-on: Create peak summit file

1. Compute an expression on every row Tool: toolshed.g2.bx.psu.edu/repos/devteam/column_maker/Add_a_column1/1.3.0 with the following parameters:
• “Add expression: c2+c5
• “as a new column to”: our peak file Peak regions (the interval format file)
• “Round result?”: YES

This will create an 8th column in our table, which we will use in our next step:

1. Rename the output Peak regions new column

2. Compute an expression on every row Tool: toolshed.g2.bx.psu.edu/repos/devteam/column_maker/Add_a_column1/1.3.0 rerun this tool on the last result with:
• “Add expression: c8+1
• “as a new column to”: the Peak regions new column file we just created
• “Round result?”: YES
3. Rename this file Peak summit regions

Now we cut out just the chromosome plus the start and end of the summit:

### hands_on Hands-on: Cut out columns

1. Cut Tool: cut1 columns from a table with the following settings:
• “Cut columns”: c1,c8,c9
• “Delimited by Tab”: Tab
• “From”: Peak summit regions

The output from Cut will be in tabular format.

2. Change the format to interval (use the galaxy-pencil) since that’s what the tool Intersect expects.

### tip Tip: Changing the datatype

• Click on the galaxy-pencil pencil icon for the dataset to edit its attributes
• In the central panel, click on the galaxy-chart-select-data Datatypes tab on the top
• Select interval
• Click the Change datatype button

The output should look like below:

## Get gene names

The RefSeq genes we downloaded from UCSC did only contain the RefSeq identifiers, but not the gene names. To get a list of gene names in the end, we use another BED file from the Data Libraries.

There are several ways to get the gene names in, if you need to do it yourself. One way is to retrieve a mapping through Biomart and then join the two files (Join two Datasets side by side on a specified field tool). Another is to get the full RefSeq table from UCSC and manually convert it to BED format.

1. Import Tool: upload1 mm9.RefSeq_genes_from_UCSC.bed from Zenodo or from the data library:

https://zenodo.org/record/1025586/files/mm9.RefSeq_genes_from_UCSC.bed

• Open the Galaxy Upload Manager (galaxy-upload on the top-right of the tool panel)

• Select Paste/Fetch Data
• Paste the link into the text field

• Change Genome to mm9

• Press Start

• Close the window

By default, Galaxy uses the URL as the name, so rename the files with a more useful name.

### tip Tip: Importing data from a data library

As an alternative to uploading the data from a URL or your computer, the files may also have been made available from a shared data library:

• Go into Shared data (top panel) then Data libraries

• Click on “Training data” and then “Introduction - From peaks to genes”

• Select the desired files
• Click on the To History button near the top and select as Datasets from the dropdown menu
• In the pop-up window, select the history you want to import the files to (or create a new one)
• Click on Import

As default, Galaxy takes the link as name, so rename them.

2. Inspect the file content to check if it contains gene names. It should look similar to below:

3. Rename it mm9.RefSeq_genes
4. Apply the tag #genes

## Repeat workflow

It’s time to reuse the workflow we created earlier.

### hands_on Hands-on: Run a workflow

2. Find the workflow you made in the previous section, and select the option Run
3. Choose as inputs our mm9.RefSeq_genes (#genes) BED file and the result of the Cut tool (#peaks)
4. Click Run workflow

The outputs should appear in the history but it might take some time until they are finished.

We used our workflow to rerun our analysis with the peak summits. The Group tool again produced a list containing the number of genes found in each chromosome. But wouldn’t it be more interesting to know the number of peaks in each unique gene? Let’s rerun the workflow with different settings!

### hands_on Hands-on: Run a workflow with changed settings

2. Find the workflow you made in the previous section, and select the option Run
3. Choose as inputs our mm9.RefSeq_genes (#genes) BED file and the result of the Cut tool (#peaks)
4. Click on the title of the tool Group tool to expand the options.
5. Change the following settings by clicking on the galaxy-pencil (pencil) icon on the left:
• “Group by column”: 7
• In “Operation”:
• “On column”: 7
6. Click Run workflow

Congratulations! You should have a file with all the unique gene names and a count on how many peaks they contained.

### question Questions

The list of unique genes is not sorted. Try to sort it on your own!

### solution Solution

You can use the tool “Sort data in ascending or descending order” on column 2 and “fast numeric sort”.

One of the most important features of Galaxy comes at the end of an analysis. When you have published striking findings, it is important that other researchers are able to reproduce your in-silico experiment. Galaxy enables users to easily share their workflows and histories with others.

To share a history, click on the galaxy-gear (gear) symbol in the history pane and select Share or Publish. On this page you can do 3 things:

This generates a link that you can give out to others. Anybody with this link will be able to view your history.

2. Publish History

This will not only create a link, but will also publish your history. This means your history will be listed under Shared Data → Published Histories in the top menu.

3. Share with Individual Users

This will share the history only with specific users on the Galaxy instance.

### hands_on Hands-on: Share history and workflow

2. See if you can do the same with your workflow!
3. Find the history and/or workflow shared by your neighbour

Histories shared with specific users can be accessed by those users in their history menu (galaxy-gear (gear) icon) under Histories shared with me.

# Conclusion

trophy You have just performed your first analysis in Galaxy. You also created a workflow from your analysis so you can easily repeat the exact same analysis on other datasets. Additionally you shared your results and methods with others.

### keypoints Key points

• Galaxy provides an easy-to-use graphical user interface for often complex commandline tools

• Galaxy keeps a full record of your analysis in a history

• Workflows enable you to repeat your analysis on different data

• Galaxy can connect to external sources for data import and visualization purposes

• Galaxy provides ways to share your results and methods with others

# References

1. Li, X., L. Li, R. Pandey, J. S. Byun, K. Gardner et al., 2012 The Histone Acetyltransferase MOF Is a Key Regulator of the Embryonic Stem Cell Core Transcriptional Network. Cell Stem Cell 11: 163–178. 10.1016/j.stem.2012.04.023

# Feedback

Did you use this material as an instructor? Feel free to give us feedback on how it went.

# Citing this Tutorial

1. Anne Pajon, Clemens Blank, Bérénice Batut, Björn Grüning, Nicola Soranzo, Dilmurat Yusuf, Sarah Peter, Helena Rasche, 2020 From peaks to genes (Galaxy Training Materials). /archive/2020-12-01/topics/introduction/tutorials/galaxy-intro-peaks2genes/tutorial.html Online; accessed TODAY
2. Batut et al., 2018 Community-Driven Data Analysis Training for Biology Cell Systems 10.1016/j.cels.2018.05.012

### details BibTeX

@misc{introduction-galaxy-intro-peaks2genes,
author = "Anne Pajon and Clemens Blank and Bérénice Batut and Björn Grüning and Nicola Soranzo and Dilmurat Yusuf and Sarah Peter and Helena Rasche",
title = "From peaks to genes (Galaxy Training Materials)",
year = "2020",
month = "12",
day = "01"
url = "\url{/archive/2020-12-01/topics/introduction/tutorials/galaxy-intro-peaks2genes/tutorial.html}",
note = "[Online; accessed TODAY]"
}
@article{Batut_2018,
doi = {10.1016/j.cels.2018.05.012},
url = {https://doi.org/10.1016%2Fj.cels.2018.05.012},
year = 2018,
month = {jun},
publisher = {Elsevier {BV}},
volume = {6},
number = {6},
pages = {752--758.e1},
author = {B{\'{e}}r{\'{e}}nice Batut and Saskia Hiltemann and Andrea Bagnacani and Dannon Baker and Vivek Bhardwaj and Clemens Blank and Anthony Bretaudeau and Loraine Brillet-Gu{\'{e}}guen and Martin {\v{C}}ech and John Chilton and Dave Clements and Olivia Doppelt-Azeroual and Anika Erxleben and Mallory Ann Freeberg and Simon Gladman and Youri Hoogstrate and Hans-Rudolf Hotz and Torsten Houwaart and Pratik Jagtap and Delphine Larivi{\{e}}re and Gildas Le Corguill{\'{e}} and Thomas Manke and Fabien Mareuil and Fidel Ram{\'{\i}}rez and Devon Ryan and Florian Christoph Sigloch and Nicola Soranzo and Joachim Wolff and Pavankumar Videm and Markus Wolfien and Aisanjiang Wubuli and Dilmurat Yusuf and James Taylor and Rolf Backofen and Anton Nekrutenko and Björn Grüning},
title = {Community-Driven Data Analysis Training for Biology},
journal = {Cell Systems}
}
`