KNIME Nodes

Learning Objective
You will learn how to import new applications written in SeqAn into the KNIME Eclipse plugin. After completing this tutorial, you will be able to use self made applications in KNIME workflows.
Difficulty
Basic
Duration
1.5 h
Prerequisites
A First Example, Parsing Command Line Arguments

In this tutorial you will learn how to integrate new apps written in SeqAn into a KNIME workflow. The first part consists of preparing a dummy app such that it can be used in a KNIME workflow and in the second part you are asked to adapt the app such that it becomes a simple quality control tool.

Important

The steps described here are necessary if you want to develop and test new SeqAn apps in KNIME. If you only want to use existing SeqAn apps in KNIME follow Creating Workflows with KNIME.

Importing SeqAn apps into KNIME

For the first part of the tutorial follow the instructions in Generating SeqAn KNIME Nodes and import a dummy SeqAn app into KNIME.

Create a useful KNIME workflow

In the second part of the tutorial you are asked to modify the app you imported into KNIME such that it becomes a quality control tool.

Type
Transfer
Objective
Create a simple read mapping workflow in KNIME using ‘razers3’ and map the reads in reads.fastq to ref.fasta. Configure the node to use a percent-identity value of 99 and the output format could be ‘razers’. The third to last and second to last column show the matching position in the reference (begin and end respectively). The last one represents the number of matching characters in percent.

You probably observe that you do not find a lot of matches. The reason for this are incorrectly called bases at the end of the reads.

Type
Transfer
Objective
Modify the ‘knime_node’ app such that it becomes a quality trimmer. You might start by just deleting the last bases of the reads (say 4 or 5) and then make the cutting depending on the actual quality values. Include the node into you workflow and inspect if the results change.

Tip

KNIME needs to know the input and output ports of a node. Therefore we must specify them using ArgParseArgument::INPUTFILE or ArgParseArgument::OUTPUTFILE as can be seen in the ‘knime_node’ app. In addition, KNIME needs to know the valid file endings, which you can specify with setValidValues, which is also shown in the example.

comments powered by Disqus