Describe gene mapping differences across methods

[sr320]

look at reports

[mgavery]

Questions:

1. Are the fragment (insert) sizes the same for all libraries? How does bismark (bowtie) get the information about insert size, can't this affect mapping?

2. Is read quality different - this should be obvious from reports

3. Run fastqc on the trimmed reports, this should help compare

[kubu4]

Are the fragment (insert) sizes the same for all libraries? How does bismark (bowtie) get the information about insert size, can't this affect mapping?

Is read quality different - this should be obvious from reports

Run fastqc on the trimmed reports, this should help compare

Already done:

MultiQC report for RRBS/WGBS:

https://gannet.fish.washington.edu/Atumefaciens/20200305_methcompare_fastp_trimming/multiqc_report.html

[mgavery]

Why is MBD-BS-Seq not mapping as well? -one possibility is that symbiont DNA is highly methylated and is getting preferentially pulled down relative to host DNA for this method.
-test this by mapping to some type of in silico symbiont genome

[mgavery]

symbiodinium genome and methylation references: https://www.nature.com/articles/s42003-018-0098-3

[kubu4]

Consider sorting reads taxonomically before mapping?

[mgavery]

How do you do that?

[kubu4]

I've done this using MEGAN6 to separate dinoflagellates from crab sequences.

[kubu4]

Basically, do some BLASTing and InterProScan stuff and then the software assigns taxonomies to individual reads, then allows you to pull reads at any level of taxonomic hierarchy.

[hputnam]

We were able to use Sym C1 for tests of alignments in the Pacuta data.