Please enable JavaScript.

Coggle requires JavaScript to display documents.

High Throughout Sequencing - Coggle Diagram

- - - - Made up of 3 parts: 1. sequencer binding site (hybridization region), 2. index (barcode), 3. primer binding site
      - single indexing if it is just on one site (5' end), dual indexing if adaptors are added to both ends of target DNA
    - - 4-scan: slowest part of the sequencing process, every hybridization reaction had to be scanned in 4 different wavelengths to determine if it was A, T, G or C that bound
      - 2-scan: Rather than scan each hybridization reaction 4 times, it is scanned 2 times. First scan is red, if reaction, it is C. Second scan green, if reaction it is T. If reaction occurs on both scans, it is an A. If no reaction occurs, it is a G.
    - - Every time PCR is done it adds bias to data
      - short-reads pose problems in scaffolding, similar sequences and gene copies create difficulty when deciding structure
  - - - feeds template through a transmembrane protein, passing through a membrane with a high electrical charge. electric signal is recorded as molecules pass through nanopore, electric singal is different for A,T,C and G. THis is because A, T, G and C carry different electrical charges
      - higher error rate than PacBio
      - Doesnt use flourescence!
      - more portable! Easy field tool.
      - Can sequence RNA directly!
    - - holds a single molecule in place, template is generation by adding nucleotides (phospholinked nucleotides)
        uses flouresence like illumina
      - differences from Illumina: Illumina when a nucleotide is added, we do not add the next nucleotide BEFORE we scan. PacBio REMOVES the flouresecent molecules as it binds, making it more RT. Also polymerase is held in place, not the DNA strand like in Illumina
      - error rates are RANDOM, so if you sequence the same molecule 10 times, you can take the consensus of those reads and rely on it being pretty accurate, as errors wont occur in the same spot in a pattern
      - HiFi = High Fidelity Reads
      - You can tell something is methylated if the sequencing colors take a long time to show up between bases. The spaces inidicate methylated areas
      - Two modes:
        
        CCS : Circular Consensus Sequencing - Sequencing a single molecule many times resulting in multiple reads, error rate 99%
        
        CLR: Continuous Long Read Sequencing - sequencing many molecules just one time each, error rate much higher (90%) but much longer reads
    - - Main differences compared to second gen: it is long read and single molecule
      - direct molecules sequencing, there is not amplification necessary
      - can sequenc RNA directly
      - DNA modifications can be captured: metylation, etc
      - FAST
  - - - Amplification (PCR?) during sequencing
      - High read accuracy (often 99%)
      - Works with almost any DNA sample
      - Low amount of input DNA needed
      - Fragmented DNA
      - Cheaper
    - - No amplification needed (no PCR bias)
      - Low single read accuracy/high consensus accuracy (random errors dont occur in the same spots)
      - High quality DNA needed
      - High amount of input DNA needed
      - DNA fragments at least 40kb long
      - Expensive
      - Paralogous sequences like chromosome areas are much easier to map out using long reads, just look like repeats on short reads
- - - - LncRNAs: Long non-coding RNA, >500nt
- - - - FIsher's Exact Test
        
        cutoffs for significance are somewhat arbitrary
    - - http://david.ncifcrf.gov/
      - http://pantherdb.org/
        
        compares gene list with a reference
  - - - tool to trim out artefacts in sequencing data
    - - tool to use with sam files
    - - alignment tool for mapping NGS reads
    - - quality check sequencing data
    - - used to calculate gene counts
  - - - consists of two steps: 1. all pair-wise distances between objects (genes or chips) are calculated (dist). 2. the dendogram is then drawn using distances by a selected method
- - - - Protein kinase (molecular function)
      - Apoptosis (a biological process)
      - Membrane (location)
      - One gene can be linked to multiple GO terms
      - most GO terms are created automatically, whch is good to keep in mind (they could be wrong!)
- - - - Sequence data
        
        Two lines, first line starts with >header, data type (DNA, RNA, mRNA etc), second line is sequence
    - - Sequence data including quality
        
        four lines, header starts with , second line is sequence, quality header starts with +, fourth line contains the quality of each base in the second line
    - - general feature format files. Current format is GFF3 and is based on version 3. Each file contains 9 columns of fata, plus optional track definition lines. GTF is identical to GFF version 2.5, http://gmod.org/wiki/GFF3
    - - aligned/mapped data
        
        sequence alignment/map formap. Header can be accessed using samtools view -H. Alignment records can be accessed using samtools view; use -h to inlcude Header, http://samtools.github.io/hts-specs/SAMv1.pdf
    - - variant calling
        
        only contains variants in a sequence. http://samtools.github.io/hts-specs/VCFv4.2.pdf, http://vcftools.sourceforge.net/VCF-poster.pdf