Buffl
Buffl
Genome Analysis

12 - NGS

MF
by Mista F.

Briefly describe the process of Sanger sequencing.

  1. DNA polymerase -> fluorescent terminators

  2. results in random fragments with marked terminator

  3. sort fragments by length (gel-electrophoresis)

  4. computer reads light signals -> color at each base

—> But:

  • short reads

  • high error rate

  • complicated, takes long


Name some Next-generation-sequencing approaches.

  • 2nd Generation (collection of molecules)

    • Illumina

    • Roche/454

    • SOLiD

  • 3rd Generation (single molecules)

    • Oxford Nanopore

    • Pacific Biosciences

    • Ion Torrent


Give an overview of the experimental step in an RNA-sequencing protocol.

  1. sample extraction

  2. target enrichment

  3. cDNA (add adapters for PCR & amplify)

  4. Sequencing

  5. Mapping

  6. Data analysis


What is the general workflow of Illumina sequencing?

  1. Library preparation

    • Fragment

    • add adaptors

    • amplify

  2. Cluster generation

    • attach to flow cell

    • bridge amplification -> build double strand from fragment

    • split bridge into two single strands

    • generate clusters

  3. Sequencing (by synthesis)

    • add fluorescent nucleotides

    • laser -> image -> determine base


Name different general 2nd Gen NGS approaches.

  • sequence-by-synthesis

    • Illumina

  • sequence-by-ligation

    • SOLiD

  • pyrosequencing

    • 454


What is a measure of base quality in NGS reads?

  • PHRED quality score

  • embedded into FASTQ file

  • defined as estimated error probability


Name a tool used to address the quality of NGS reads. How is quality improved/checked?

  • FastQC

  • quality control:

    • trimming (Trimmomatic)

      • fixed length

      • adaptive (based on quality score)


Explain the FASTQ format.

  • extension to FASTA

  • includes PHRED quality scores (from 0 to 93)

    • encoded in ASCII 33–126

    • range of error probability

      • from 1 (wrong base)

      • to 10^-9.3 (extremely accurate)

  • Format:

    • sequence header = @ID name length=n

    • sequence

    • quality header = +ID name length=n

    • PHRED scores


What are some common sequence artifacts in NGS data?

  • read errors

  • base call errors

  • small insertions/deletions

  • poor quality reads

  • primer/adapter contamination —> trim adapters


What is a contig?

  • contig = contiguous sequence

    • longest overlap between reads


Describe how a genome is assembled.

  1. Fragment DNA

  2. Find overlaps

  3. merge overlaps into contigs

  4. merge contigs into scaffolds


Why are genomes hard to assemble?

  • Accuracy

    • no ground truth

  • Sequencing errors

  • Computationally expensive

  • Differentiating biology


How does high read coverage help in sequencing? Which problem cannot be overcome with it?

  • decreases error rates

  • problem: repeats


What graph-based solution for assembly would you use for long and short reads? How are repeat sequence represented in the graphs?

  • Long reads -> overlap graph

  • Short reads -> de Brujin graph (3-mers only)

-> Repeat sequences create a fork in graphs


Describe a typical RNA-seq workflow.

  1. Quality control

  2. Read alignment to reference genome

  3. Transcriptome assembly

  4. Differential expression


Name the three different transcriptome assembly strategies.

  • reference-based

  • de novo

  • combined


Name some splice-aware aligners. What tool can be used to create and traverse overlap graphs?

aligners:

  • TopHat

  • Blat

graph:

  • Cufflinks


How do you normalize read counts for genes based on read numbers and length? What is the benefit?

  • How? —> Raw reads/length = normalized reads

  • Why? —> compare different conditions/experiments


Explain some measures to normalize transcript expression.

  • RPKM: reads per kilobase of transcript per million mapped reads

    • norms by:

      • read length

      • total number of mapped reads

  • FPKM: fragments per kilobase of transcript per million mapped reads

    • only for paired-end

    • RPKM with dependency of paired read

  • TPM: transcripts per million

    • norms by:

      • first: gene length

      • then: sequencing depth


Descrive the VCF file format.

—> Variant Call Format

-> describe variations of sequences in different conditions

FORMAT column:

-> homozygous = same copy

-> heterozygous = different copy

What is single-cell sequencing? What are the main differences to bulk RNA-seq?

  • single-cell:

    • use barcodes to identify RNA of individual cells

  • differences:

    • bulk RNA-seq:

      • comparative transcriptomics

      • disease biomarkers

      • homogenous systems

      • ~20.000 mRNA transcripts

    • scRNA-seq:

      • identify rare cell populations

      • cell population dynamics

      • define heterogeneity

      • 200-10.000 mRNA transcripts/cell


What are common applications for single-cell RNA sequencing?

  • Deconvolution -> cluster cell populations

  • Trajectory -> cell differentiation

  • Networks -> Gene Regulatory Networks (GRNs)


Describe the steps of Drop-seq.



Join Course
Preview

Author

Mista F.

Information

Last changed

Report course
© 2023 Buffl GmbH