Please enable JavaScript.

Coggle requires JavaScript to display documents.

Advanced transcriptome/metagenome assemblies, Advanced Assemblies - Coggle…

- - - - Program - Trinity - first pass greedy assembly, followed by clustering resulting contigs
      - Assembly each cluster's contigs using de Bruijn graph
        
        Compact each de Bruijn graph (in each cluster)
        
        Annotate the compacted graph with read depth (in each cluster)
        
        Join contigs further using the read depth and read pairing information - generates the full transcripts (full sequences for each cluster)
- - - - If not okay, go back to collect more samples
      - Option 1: De novo assemble the reads and identify contigs
        
        Assembly using - Megahit, SPAdes, IDBA, Minia3
        
        Identification
        
        Assuming contigs belonging to a genome would have more or less the same read coverage - identify them as belonging to one genome
        
        Assuming contigs belonging to a genome would have similar G-C content - identify them as belonging to one genome
        
        Identify contigs similar to a relevant reference genome
      - Option 2: Map the reads to a related/target reference and just assemble
        
        Accessory genome differences won't be mapped - regions present in the strain but not in reference won't be included
        
        Strains that are similar but separate may also get mapped to the reference - resulting in an assembly not belonging to a single strain
      - To get the metabolic capabilities of the community or target strains
        
        Assemble and then annotate the metabolic pathways
        
        Tool - Prokka - can annotate part genes too (even the assembly is slightly incomplete due to read coverage issues)
        
        Uses BLAST to search various databases and use the hits (based on an E value threshold) to annotate the queries.
        
        Only for Bacteria, Archaea and Viruses
        
        Workshop 9
        
        .gff files
        
        Lecture 19, slide 27-29
      - Quality metric - N50 (end of workshop 8)
- - - - Systematic errors - incorrect count of specific bases (called homopolymer errors)
      - Random errors - random substitutions
      - Tool: Circlator to trim contigs - removes overlap artifacts and orientates the chromosome sequence
    - - Systematic error - lower depth in high GC content regions
      - Random error - homopolymer errors
    - - Homopolymer errors/indels - frameshift mutations and broken genes
      - SNP/substitution errors - missense or nonsense mutation
      - Defunct sequencing technologies - Roche/454, Ion Torrent