Mick at 16:50, 13 January 2026

2026-01-13T16:50:56Z

← Older revision		Revision as of 18:50, 13 January 2026
Line 122:		Line 122:
	<hr>		<hr>

	<p>You can find the answers [[Postprocess_exercise_answers here]]</p>		<p>You can find the answers [[Postprocess_exercise_answers\|here]]</p>

	<p><b>Congratulations—you have completed the exercise!</b></p>		<p><b>Congratulations—you have completed the exercise!</b></p>

Mick at 16:50, 13 January 2026

2026-01-13T16:50:20Z

← Older revision		Revision as of 18:50, 13 January 2026
Line 122:		Line 122:
	<hr>		<hr>

	<p>You can find the answers here~~: <a href="Postprocess_exercise_answers">Postprocess_exercise_answers</a>~~</p>		<p>You can find the answers [[Postprocess_exercise_answers here]]</p>

	<p><b>Congratulations—you have completed the exercise!</b></p>		<p><b>Congratulations—you have completed the exercise!</b></p>

Mick at 09:13, 7 January 2026

2026-01-07T09:13:03Z

← Older revision		Revision as of 11:13, 7 January 2026
Line 34:		Line 34:
	java -jar /home/ctools/picard_2.23.8/picard.jar MarkDuplicates \		java -jar /home/ctools/picard_2.23.8/picard.jar MarkDuplicates \
	-I [input.bam] \		-I [input.bam] \
	-M [metrics.txt] \		-M [metrics.txt (this is part of the output, call it what you wish)] \
	-O [output.bam]		-O [output.bam]
	</pre>		</pre>

Mick at 13:52, 20 November 2025

2025-11-20T13:52:56Z

Show changes

WikiSysop: Created page with "

Overview

First:

Navigate to your home directory:
Create a directory called "postalign"
Navigate to the directory you just created.

In this exercise, we will pre-process bam-files so they are ready for SNP calling. This is necessary to reduce the high number of potential false SNPs that will get called. You will try to:

Mark read duplicates from the BAM-files
Merge BAM files

Duplicate removal

We are..."

2024-03-19T15:29:09Z

Created page with "<H2>Overview</H2> First: <OL> <LI>Navigate to your home directory: <LI>Create a directory called "postalign" <LI>Navigate to the directory you just created. </OL> In this exercise, we will pre-process bam-files so they are ready for SNP calling. This is necessary to reduce the high number of potential false SNPs that will get called. You will try to: <OL> <LI>Mark read duplicates from the BAM-files <LI>Merge BAM files </OL> <H2>Duplicate removal</H2> <p>We are..."

New page

<H2>Overview</H2>

First:
<OL>
<LI>Navigate to your home directory:
<LI>Create a directory called "postalign"
<LI>Navigate to the directory you just created.
</OL>

In this exercise, we will pre-process bam-files so they are ready for SNP calling. This is necessary to reduce the high number of potential false SNPs that will get called. You will try to:
<OL>
<LI>Mark read duplicates from the BAM-files
<LI>Merge BAM files
</OL>

<H2>Duplicate removal</H2>

<p>We are going to work on data from a Han Chinese individual (HG00418) that was sequenced to around ~40X using Illumina paired-end sequencing. We are not going to use all the data but only some data from two libraries that are both paired-end and we are only going to use chr20 instead of the entire human genome.</p>

The first Library can be found here:
<pre>
/home/projects/22126_NGS/exercises/dupremoval/ERR016028_chr20_sort.bam
</pre>

We have already trimmed and aligned the data. As the filename indicates, the file has been sorted.

We will mark duplicates them using Picard:
<pre>
java -jar /home/ctools/picard_2.23.8/picard.jar MarkDuplicates -I [input BAM] -M [metrics output text file] -O [output BAM file]
</pre>

Try to maybe add a suffix like "_chr20_sort_markdup.bam".

'''Q1:''' the output should indicate how many reads were marked as duplicates, how many duplicates did Picard flag?

<H3>Effect</H3>

We will look at the effect of marking duplicates, a reminder that you can inspect a specific genomic region using:

<pre>
samtools view [input.sorted.bam] [region ID]:[start]-[end]
</pre>

However, the bam file has to be indexed (and therefore sorted, however, do not worry about sorting the file again because Picard maintains the order of the reads).

Try to view the reads aligning to the following region:
<pre>
chr20:45996339-45996839
</pre>

Both for the original file and a new file you just created with duplicates marked.

'''Q2:''' why were reads ERR016028.5947720 and ERR016028.18808080 considered duplicates?

'''Q3:''' which of the two was marked as duplicate and will be ignored by genotypers? How did you determine this?

<H2>Merging reads</H2>

Very often you will have to combine different sequencing libraries together prior to genotyping.

The goal is to combine the file you just generated with the duplicates removed with another of BAM file:
<pre>
/home/projects/22126_NGS/exercises/dupremoval/ERR016025_chr20_sort_markdup.bam
</pre>

This BAM file is from the same individual, it is just a different library.

Type:
<pre>
samtools
</pre>
Find a command that can combine both files, keep them sorted and write the index out for the new file in a single command.

'''Q4:''' which commander can do this?

Run the command that you found and use the options:

<pre>
-c --write-index
</pre>

The first option is to keep the re groups as they are is and the second is to simultaneously generate the index. The new file should be named '''HG00418_chr20_sort_markdup.bam'''. Inspect the new file using '''samtools view''' and '''less'''

'''Q5:''' which field allows you to determine the sample of origin (i.e. from which file was a read from)?

'''Q6''' What is the term for pooling multiple samples into a single sequencing run?

'''Q7''' What is the term for the computational process where we take the raw data from a single sequencing run and separate them into different samples?

Please find answers [[Postprocess_exercise_answers|here]]

'''Congratulations you finished the exercise!'''

<p></p>
<p></p>
<p></p>

Postprocess exercise - Revision history

Mick at 16:50, 13 January 2026

Mick at 16:50, 13 January 2026

Mick at 09:13, 7 January 2026

Mick at 13:52, 20 November 2025

WikiSysop: Created page with "

Overview

Duplicate removal