What is the difference between k-mer size used for building local and global assembly graph?

Question

[VIJITH KUMAR]

I have recently started working on a bioinformatics project. My project is to perform analysis of "Whole Genome Sequencing" data. The read files - R1 and R2 - measure 30GB each. We are using an IBM server machine for this purpose, and "Universal Sequencing Tell-Seq software", in order to perform the analysis. The IBM server has got an "Intel 8-core processor", and 197 GB RAM. As of now, I have provided the following wrapper script:

 run_tellink.sh \
  -r1 /path/to/R1/data 
  -r2 /path/to/R2/data 
  -i1 /path/to/I1/data 
  -o <path/to/output> 
  -k <k-mer size selected to build global assembly graph> 
  -lc <local k-mer size selected to build local assembly graph> 
  -p <prefix name>

Here, I would like to know what is the difference between the two different k-mer size - the one used for building the global assembly graph and the one used for building the local assembly graph. Can someone provide me an information or resources to gain some understanding about it?

[Ram]

You should contact Universal Sequencing with your questions - you're running their wrapper so they'd know best about their parameters. Read the documents under the "software user guides" section before you contact them so you get some context on your questions: https://www.universalsequencing.com/user-guides

From your post, it does not look like you have a lot of bioinformatics experience. Try and understand basics on graph based assembly as well as local realignment - I don't know the former (and don't have a good understanding of the latter) either, but it looks like a concept key to resolving your question.