From Raw Sequences to Assembled Genomes

Library Construction

• involves ligating known DNA sequences (adapters) onto unknown genomic fragments, so that the fragments can bind to the flow cell, be amplified and sequenced from both ends

• terminal sequences: complementary to oligos on flow cell; allow DNA fragments to attach (tether) to sequencing surface

• adapters: short know DNA sequences ligated via enzymes, enables amplification and sequencing

• primer binding sites: where sequencing primers bind, required to initiate DNA synthesis

• unknown DNA sequence: can obtain ~150-250 bp from the ends of each fragment

Library Construction Figure

Why Sequence Both Ends?

• helps resolve repetitive DNA

• places fragments more accurately in the genome

Cluster Generation

• steps: 2) generate clusters of amplified DNA sequences

• DNA strands bind to flow cell via base pairing to 1st terminal sequence

• flow cell oligos are extended using DNA polymerase

• 3’ ends of the extended sequences hybridize w/ nearby oligos

• flow cell oligos are extended using DNA polymerase (extension, bridge formation)

  • repeat to generate clusters

Remove Reverse Strands

• after cluster generation, each cluster contains forward + reverse strands but sequencing-by-synthesis method requires a single stranded template, all oriented the same way

  • one strand must be removed

• restriction sites are included into the adapter; the site is only present in one strand orientation

• restriction enzyme cuts and removes the reverse strands

Sequencing by Synthesis

• steps: 3) massively parallel sequencing of clusters

• nucleotide analogs w/ cleavable fluorophore and terminating groups are added and then terminator/dye are cleaved each cycle

How is the Genome Divided?

• into coding regions (exome) and non coding regions

  • exons: differential splicing can occur

• transcriptome: complete set of RNA transcripts produced in a cell (or tissue) at a given time and condition

  • can be used to compare differences between cells

  • tells us which genes/region of genome is transcribed

    • ome ending: studying all at once

• enhancers: regions that affects the activity of a promoter (can be far from gene)

• promoters: where core RNA polymerase machinery binds to initiate transcription (has to be close to gene)

• mutations in regulatory non-coding regions can affect gene expression

• mutations in exons can affect the underlying amino sequence as well as gene expression

How is the Genome Divided? FIGURE

What does Exome Sequencing Identify?

• exome sequencing identifies causal mutations even in a small patient pool

• mutations in embryonic myosin MYH3 are thought to cause Freeman-Sheldon syndrome, a severe congenital joint disorder

  • NS/SS/I = potential damaging mutation

• exome sequencing: MYH3 was the only damaged coding region in all 4 affected patients and not found in a public database of common SNPs, or in 8 control individuals

• figure: as you compare patient genome and reference genome, you narrow down which gene is the problem (where the mutation is)

What does Whole Genome Sequencing Reveal?

• whole genome sequencing reveals the role of non-coding mutations

• mutations affecting enhancers, promoters, silencers, insulator elements can have substantial affects on gene expression, and cannot be captured by exome sequencing

Conclusions so Far

• the development of new technologies have dramatically lowered the cost of DNA sequencing

• number of genes hasn’t increased that much in more complex organisms – complexity during animal evolution must be driven by increasingly sophisticated gene regulation

• the ability to obtain massive amounts of sequencing data have helped to discover the causes of many rare genetic syndromes (even w/ very few patients)

Why has it taken ~20 years to complete the Human Genome Project?

Going from raw sequences to an assembled genome is not simple

• putting short reads in the correct order requires there to be sufficient read coverage

• sequenced genome may have deletions, insertions, rearrangements, substitutions, etc. compared to a reference genome

• genomes have many regions that are highly repetitive

From Reads to Contigs

• contig: a consensus sequence that is result of assembling together overlapping fragments

• polymorphism: common DNA sequence variation at a specific genomic position that exists in a population

  • alleles are different versions at that site

  • typically small differences (single nucleotide polymorphisms, small indels)

• coverage: the number of times each position is sequenced in an experiment; poor coverage can lead to assembly errors

• major challenge for sequencing is repetitive DNA

Repetitive Elements

• make up the majority of human genome (66-69% is repetitive or derived from repeats)

Major types:

• tandem repeats: short repeats including minisatellites (10-60 bps) found in centromeres, microsatellites (2-8 bps) found in telomeres

• interspaced repeats: transposable elements

  • found throughout genome, enriched in centromeres, pericentric/intercalary heterochromatin

Satellite DNA Sedimentation

• satellite DNA differentially sediments in a CsCl gradient due to different base composition than the “main band” DNA

• don’t have same AGCT representation

• satellite bands are repetitive DNA that don’t co-sediment w/ main band

• AT DNA is less dense than GC DNA

Transposable Elements in Maize

• coloration of maize kernels are caused by the disruption of a pigment gene by a transposable element Dissociation (Ds)

  • if left in pigment gene → purple corn (WT)

• dissociation transposition (jumping) requires an enzyme produced by a second transposable element Activator (ac)

• this showed that the genome is much mroe dynamic than we previously thought

  • genes can literally jump out from and into genomes

Transposons Definitions

• DNA transposons replicate by cutting and pasting themselves into different parts of the genome

• LTR (long terminal repeat) retrotransposons: transcribed into RNA and reverse-transcribed into dsDNA in the cytoplasm using a tRNA primer (copy and paste)

• Non-LTR retrotransposons: transcribed into RNA which is reverse transcribed into DNA used nicked host DNA as a primer in the nuclease (copy and paste)

• both non-LTR and LTR use RNA intermediates → reverse-transcribe to DNA

Transposable Elements in our Genome

• transposable elements make up almost half our genomes

  • genomic parasites; impact on genes; important for evolution

• ~4,000 full-length LINE1 insertions and ~100,000 fragments in the human genome

  • initially thought ~40 LINE1 elements are still active in our genomes

Transposons Insertions: Diseases

• study of 240 hemophilia patients showed that 2 individuals had disease mutations caused by independent LINE1 insertions into Factor VIII

• evidence of elevated LINE1 activity in cancers, is hypothesized to play a role in mutagenizing cancer genomes to promote cancer progression

• estimated that 1 in 1,000 disease causing mutations is due to a novel LINE1 insertion

Transposons Insertions: Evolution

• new transposons insertions help to drive variation and the evolution of adaptive traits

Arc Gene

• important for many forms of neuronal plasticity (i.e memory formation)

• Arc = activity regulated cytoskeleton

• related to repetitive DNA b/c it evolved from a retrotransposon (type of repetitive DNA)

Arc1

• Arc1 encodes a retroviral-like protein that traffics b/w neurons

  • independent domestication of retroelements happened many times in evolution (suggests strong selective advantage0

• Arc1 protein is made in muscle cells via neuronal Arc mRNA

  • neurons produce Arc mRNA

• Arc1 forms viral-like capsids in purified exosomes

  • these particles encapsulate Arc mRNA and are released in extracellular vesicles (exosomes)

• our ability to form new memories requires a gene that evolved from a retro-element

  • loss of Arc → severe memory deficits

Why can identical repetitive DNA sequences have different biological effects?

• even if the 2 DNA sequences are identical at the nucleotide level, their biological effect can be completely different depending on where they are in the geome

  • sequence ≠ function by itself

• eg. different contexts: different neighboring genes, regulatory elements, cell type

• effects can be very different depending on where they land

  • eg. lands in promoter → alters transcription; exon → disrupts protein; intron → affects splicing

  • in Arc: one copy of the repetitive retrotransposon landed in the right genomic context and became essential for memory

How does Paired-End Sequencing Generate Reads

Paired reads are produced thru sequential sequencing of forward and reverse strands

• DNA is fragmented to segments of known length, adapters are ligated to both ends and contain sequencing primer binding sites

  • this defines the distnaces between reads

• cluster generation (contain forward and reverse strands)

• Read 1: reverse strands are cleaved, leaving single-stranded forward templates

  • forward strands are sequenced (first paired read; forward read sequenced)

  • sequencing primer binds to the adapter

• Read 2: forward strand is removed, reverse strand is regenerated by adding dNTPs and DNA polymerase

  • sequencing primer binds on the other adapter

  • reverse read sequenced (2nd paired read)

• final results: reads from opposite ends/orientation, but from the same fragment and separated by a known insert size

How does Paired-End Sequencing Generate Reads FIGURE

• within the two adapters that are ligated onto either end of DNA molecules are sequencing primer binding sites

• paired end sequencing can help us to map some repetitive sequences

Paired End Sequencing Mapping

Sanger Sequencing Elegant Approach

• increase the read length: get sequencing reactions to approach the size of chromosomes

• long read sequencing analyzes single DNA molecules

• long read sequencing has higher error rate (0.2-3% vs 0.1%) and is more expensive than short read sequencing

• short and long read sequencing using all platforms can be (and are frequently) combined

  • this approach led to the full sequencing of human chromosomes

PacBio Sequencing

• uses nanostructure arrays borrowing tech developed for semiconductors

  • “zero-mode waveguide nanostructured arrays”; tiny hole in a meal film, light can only penetrate a few nanometers into the well; only fluorophores right at the polymerase active site are illuminated

  • tiny nanostructured wells to restrict single DNA polymerase molecules so individual nucleotide incorporation can be detected in real time

  • semiconductor allows many identical nanometer scale wells

• individual well vol = 1 zeptoliter (1 × 10^-21 liters)

  • so small that only a few molecules are present at any time

  • one polymerase dominates the signal

  • enables single-molecule detection w/o amplification

• rate of molecule diffusion is extremely fast compared to the rate of DNA base incorporation

  • when an nucleotide is incorporated, it is held by the polymerase, so the fluorophore stays long enough to be detected

PacBio Sequencing

PacBio Hifi Sequencing

• single molecule sequencing by synthesis (sequences one DNA molecule at a time)

  • no PCR amplification required

  • DNA polymerase incorporates nucleotides like normal replication and each base addition is observed in real time

• bases are labeled with dyes thru the terminal phosphate group – incorporation will lead to cleavage of the dye

Real-time Detection of Base Incorporation

• time for each base to be incorporated (miliseconds) is much faster than random diffusion (microseconds)

  • free nucleotides diffuse too quickly to produce a signal

• DNA polymerase highly processive (can synthesize >70 kilobases w/o falling off)

  • enables very long reads

• get a fluorescent spike (strong signal that lasts) for correct bases

  • DNA polymerase holds each nucleotide long enough for a detectable fluorescent signal, allowing extremely long single-molecule reads

Oxford Nanopore Sequencing

• directly measures DNA translocation

• similar to gel electrophoresis, applying a voltage gradient leads negatively charged DNA to travel towards the anode, only route is thru a pore

• translocation of DNA blocks the natural current of ions thru a pore

  • as each nucleotide passes thru the pore, the DNA partially blocks the ionic current

  • different bases cause distinct current changes

• detects DNA bases by measuring changes in ionic current as DNA molecule passes thru biological pore under applied voltage

• can detect modified bases and structural variants

Oxford Nanopore Sequencing FIGURE

Oxford Nanopore Sequencing: Shelf Life

• because nanopore sequencers use biological channels, there is a limited shelf life

  • 1 month at room temp, 3 months at 4ºC

• accessibility: data acquired in the lab via a laptop w/ a USB connection

What is the sequence of centromeric DNA?

• centromeric DNA is composed of highly repetitive sequences, making these regions difficult to sequence and assemble (“dark matter”)

• centromeres are essential for chromosome segregation and are flanked by heterochromatin

• humans: alpha satellite DNA is the main repeat in centromeres

  • centromeres organized as monomeric repeat and higher order repeats (multiple monomers in a specific order repeated many times)

  • surrounding heterochromatin contain transposable elements

• drosophila: centromeres are also highly repetitive, but differ in sequence

  • surrounding heterochromatin contain many transposable elements

What is the sequence of centromeric DNA? FIGURE

Satellite Repeats in Centromeres

• give an evolutionary advantage

• chromosomes w/ more repeats will end up inherited at a 60:40 ratio vs chromosomes w/ fewer repeats

• more repeats = stronger centromere = higher chance of being inherited

  • during female meiosis, ¼ meiotic products becomes the egg, and chromosomes w/ stronger centromeres (more repeats) are more likely to attach to the egg pole

  • centromere repeats aren’t evolving b/c they make the organism for fit, which may lead to species that are less fit (b/c centromeres are cheating in meiosis)

Satellite Repeats in Centromeres FIGURE

What did long-read sequencing reveal about fly centromeres, and how are transposable elements involved?

• long read sequencing led to the discovery that fly centromeres are filled w/ islands of retroelements (not just homogenous satellite repeats)

• centromeres are sites of evolutionary warfare since they compete for inheritance during female meiosis

  • transposons (“soldiers”) can contribute to centromere function and possibly increasing strength for centromere drive

What does comparison of human centromeres reveal about their variability and evolution?

• gapless assemblies: genome assemblies have gaps in repetitive regions

• humans are highly variable: ~50% of centromeric sequences can’t even be aligned b/w two individuals due to higher order repeat sequences

• the positions of centromeres can differ by >500 kilobases

• the size of centromeres can differ by up to 3 fold

• further evidence that human chromosome evolution is shaped by complex and often non-Mendelian forces

  • human chromosome evolution is dynamic and influence by repetitive DNA and not just coding sequences

Conclusions

• repetitive DNA has important functions in biology and disease

• the full sequencing of genomes necessitates the incorporation of long read sequencing approaches

• several sequencing approaches are combined tgt to assemble an accurate genome

• short read sequencing gives reads of the highest accuracy whereas long read sequencing enables analysis of repetitive regions