These sources provide an extensive overview of eukaryotic transcription and the history and
methodology of DNA sequencing that culminated in the complete human genome assembly.
The first source focuses on the three specialized multisubunit RNA polymerases (Pol I, Pol
II, and Pol III) in eukaryotes, detailing their distinct functions, structures, and mechanisms for
transcription initiation, elongation, and termination. The second and third sources document
the evolution of sequencing technology through three generations, from Sanger
sequencing used in the original Human Genome Project (HGP) to Next-Generation
Sequencing (NGS), and finally to Third-Generation long-read sequencing (PacBio and
Nanopore), which was essential for resolving complex, repetitive regions like centromeres and
rDNA arrays to complete the telomere-to-telomere (T2T) human genome. Both documents also
emphasize the composition and complexity of the human genome, noting that gene number is
not proportional to organism complexity and that most of the genome consists of non-coding,
repetitive DNA.
The sources provide a detailed narrative of the Human Genome Projects (HGP) and
subsequent efforts, highlighting how advances in Genome Sequencing and Assembly
technologies were necessary to tackle the size, complexity, and highly repetitive nature of
the human genome.
The Original Human Genome Project (HGP) and First-Generation Sequencing
The sequencing of the human genome officially began with the Public Human Genome
Project (HGP), which was conceived in 1984 and started in 1990. The ultimate goal was to
sequence the entire human genome, a process that initially cost approximately $3 billion
in 1990, although this cost dramatically dropped to less than $100 million within a decade
due to technological improvements, reflecting Moore's Law.
Sequencing Technology and Strategy: The HGP primarily utilized First-Generation
Sequencing (Sanger sequencing). This technology, developed in 1977, was capable of
sequencing only about 500 "letters" (base pairs) at a time. To sequence the massive
human genome, the HGP employed a specialized assembly strategy called Hierarchical
shotgun sequencing:
1. Large fragments of the human genome were cloned into Bacterial Artificial
Chromosomes (BACs), which could accommodate fragments over 300 kb.
2. These BACs were selected, ordered based on physical and genetic maps, fragmented, and
then subcloned.
3. The DNA from individual subclones was then sequenced using automated Sanger
sequencing.
This hierarchical, clone-based approach was critical because it helped avoid problems
associated with repeat sequences. In contrast, the concurrent Private Human Genome
Project (started in 1998) used a Whole Genome Shotgun approach. The sources note that
while the Whole Genome Shotgun strategy was reasonable for a draft, the sheer
pervasiveness of repetitive sequences meant it did not yield the same high-quality
reference as the HGP’s clone-based approach.
The Initial Outcome and the Reference Genome: A "complete" draft of the human
genome was achieved in 2001 by both projects, covering 83–84% of the entire genome.
The resulting Human Reference Genome (HRG), which descends from the HGP, is a
composite sequence derived from the DNA of multiple individuals (the original HGP used
thirteen anonymous volunteers). It is defined as a haploid sequence and does not
, correspond to any actual or "ideal" human individual. Even after the HGP's completion in
2003, continuous work is performed by the Genome Reference Consortium (GRC) to
correct misrepresented regions and close remaining gaps.
The Challenge of Gaps and Assembly
The initial completion in 2003 still left sequence gaps. While 99% of the euchromatic
portion was finished, two types of gaps remained:
1. Heterochromatic gaps (estimated 200 Mb), which were never intended to be
sequenced by the HGP.
2. Euchromatic gaps (24.4 Mb).
These unsequenced regions were highly problematic because large repetitive sequences
and complex allelic diversity are the two main drivers of assembly error. Shotgun
sequencing methods struggled when repetitive sequences were broken into fragments,
making reassembly difficult and often leading to the omission of parts of the repetitive
region or misconnecting pieces of chromosomes. Furthermore, approximately 5% of the
genome consists of segmental duplications (fragments 1kb to >100kb with high
sequence identity), which often led to misassembly and collapsed regions in the original
HGP assembly.
Regions that defied early assembly included:
• Alpha-satellite and centromeric transition regions.
• The short arms of acrocentric chromosomes (13, 14, 15, 21, and 22), which contain large
tracts of satellite sequences and tandem arrays of ribosomal DNA (rDNA).
Genome Re-sequencing and Second-Generation Sequencing
The advent of Second (Next)-Generation Sequencing (NGS) around 2005 marked a shift
toward massive parallel sequencing, vastly increasing the number of de novo assemblies.
NGS is characterized by "short-read" sequencing (typically 50–400 base pairs). The short
reads combined with repetitive genomes were partially managed using new assembly
algorithms based on de Bruijn graphs. However, NGS assemblies of larger genomes were
generally of poor quality when compared to the HGP's clone-based approach.
An example of a project utilizing NGS is the 1000 Genomes Project, which aimed to find
common genetic variants in diverse populations. This project sequenced 2,504 individuals
and identified 88 million variants, showing that a typical genome differs from the
reference at 4.1 to 5.0 million sites, with structural variants affecting ~20 million bases of
sequence.
The Telomere-to-Telomere (T2T) Completion and Third-Generation Sequencing
The final push to sequence the last remaining 8% of the human genome was the Telomere-
to-Telomere (T2T) completion effort. This was achieved only after technological limitations
were removed by Third-Generation Sequencing ("Long-read" sequencing), which
produces reads of several kilobases (kbs).
Key technological innovations for T2T assembly included:
• PacBio HiFi-sequencing: This provided high-accuracy long reads (e.g., 20 kb reads
with a 0.1% error rate) which were essential for assembling long, near-identical repeat
arrays. PacBio HiFi reads formed the basis of the T2T-CHM13 assembly string graph.
• Oxford Nanopore Sequencing (ONT): These ultralong reads were used to guide the
correct path ("walk") through the assembly graph when high-resolution paths were
ambiguous.
methodology of DNA sequencing that culminated in the complete human genome assembly.
The first source focuses on the three specialized multisubunit RNA polymerases (Pol I, Pol
II, and Pol III) in eukaryotes, detailing their distinct functions, structures, and mechanisms for
transcription initiation, elongation, and termination. The second and third sources document
the evolution of sequencing technology through three generations, from Sanger
sequencing used in the original Human Genome Project (HGP) to Next-Generation
Sequencing (NGS), and finally to Third-Generation long-read sequencing (PacBio and
Nanopore), which was essential for resolving complex, repetitive regions like centromeres and
rDNA arrays to complete the telomere-to-telomere (T2T) human genome. Both documents also
emphasize the composition and complexity of the human genome, noting that gene number is
not proportional to organism complexity and that most of the genome consists of non-coding,
repetitive DNA.
The sources provide a detailed narrative of the Human Genome Projects (HGP) and
subsequent efforts, highlighting how advances in Genome Sequencing and Assembly
technologies were necessary to tackle the size, complexity, and highly repetitive nature of
the human genome.
The Original Human Genome Project (HGP) and First-Generation Sequencing
The sequencing of the human genome officially began with the Public Human Genome
Project (HGP), which was conceived in 1984 and started in 1990. The ultimate goal was to
sequence the entire human genome, a process that initially cost approximately $3 billion
in 1990, although this cost dramatically dropped to less than $100 million within a decade
due to technological improvements, reflecting Moore's Law.
Sequencing Technology and Strategy: The HGP primarily utilized First-Generation
Sequencing (Sanger sequencing). This technology, developed in 1977, was capable of
sequencing only about 500 "letters" (base pairs) at a time. To sequence the massive
human genome, the HGP employed a specialized assembly strategy called Hierarchical
shotgun sequencing:
1. Large fragments of the human genome were cloned into Bacterial Artificial
Chromosomes (BACs), which could accommodate fragments over 300 kb.
2. These BACs were selected, ordered based on physical and genetic maps, fragmented, and
then subcloned.
3. The DNA from individual subclones was then sequenced using automated Sanger
sequencing.
This hierarchical, clone-based approach was critical because it helped avoid problems
associated with repeat sequences. In contrast, the concurrent Private Human Genome
Project (started in 1998) used a Whole Genome Shotgun approach. The sources note that
while the Whole Genome Shotgun strategy was reasonable for a draft, the sheer
pervasiveness of repetitive sequences meant it did not yield the same high-quality
reference as the HGP’s clone-based approach.
The Initial Outcome and the Reference Genome: A "complete" draft of the human
genome was achieved in 2001 by both projects, covering 83–84% of the entire genome.
The resulting Human Reference Genome (HRG), which descends from the HGP, is a
composite sequence derived from the DNA of multiple individuals (the original HGP used
thirteen anonymous volunteers). It is defined as a haploid sequence and does not
, correspond to any actual or "ideal" human individual. Even after the HGP's completion in
2003, continuous work is performed by the Genome Reference Consortium (GRC) to
correct misrepresented regions and close remaining gaps.
The Challenge of Gaps and Assembly
The initial completion in 2003 still left sequence gaps. While 99% of the euchromatic
portion was finished, two types of gaps remained:
1. Heterochromatic gaps (estimated 200 Mb), which were never intended to be
sequenced by the HGP.
2. Euchromatic gaps (24.4 Mb).
These unsequenced regions were highly problematic because large repetitive sequences
and complex allelic diversity are the two main drivers of assembly error. Shotgun
sequencing methods struggled when repetitive sequences were broken into fragments,
making reassembly difficult and often leading to the omission of parts of the repetitive
region or misconnecting pieces of chromosomes. Furthermore, approximately 5% of the
genome consists of segmental duplications (fragments 1kb to >100kb with high
sequence identity), which often led to misassembly and collapsed regions in the original
HGP assembly.
Regions that defied early assembly included:
• Alpha-satellite and centromeric transition regions.
• The short arms of acrocentric chromosomes (13, 14, 15, 21, and 22), which contain large
tracts of satellite sequences and tandem arrays of ribosomal DNA (rDNA).
Genome Re-sequencing and Second-Generation Sequencing
The advent of Second (Next)-Generation Sequencing (NGS) around 2005 marked a shift
toward massive parallel sequencing, vastly increasing the number of de novo assemblies.
NGS is characterized by "short-read" sequencing (typically 50–400 base pairs). The short
reads combined with repetitive genomes were partially managed using new assembly
algorithms based on de Bruijn graphs. However, NGS assemblies of larger genomes were
generally of poor quality when compared to the HGP's clone-based approach.
An example of a project utilizing NGS is the 1000 Genomes Project, which aimed to find
common genetic variants in diverse populations. This project sequenced 2,504 individuals
and identified 88 million variants, showing that a typical genome differs from the
reference at 4.1 to 5.0 million sites, with structural variants affecting ~20 million bases of
sequence.
The Telomere-to-Telomere (T2T) Completion and Third-Generation Sequencing
The final push to sequence the last remaining 8% of the human genome was the Telomere-
to-Telomere (T2T) completion effort. This was achieved only after technological limitations
were removed by Third-Generation Sequencing ("Long-read" sequencing), which
produces reads of several kilobases (kbs).
Key technological innovations for T2T assembly included:
• PacBio HiFi-sequencing: This provided high-accuracy long reads (e.g., 20 kb reads
with a 0.1% error rate) which were essential for assembling long, near-identical repeat
arrays. PacBio HiFi reads formed the basis of the T2T-CHM13 assembly string graph.
• Oxford Nanopore Sequencing (ONT): These ultralong reads were used to guide the
correct path ("walk") through the assembly graph when high-resolution paths were
ambiguous.
