Transcriptomics Lecture 3: SOLiD and Third-Generation Sequencing Technologies

Administrative Information and Exam Details\n- Next week will have no new lecture content.\n- Monday Session: A review of previous lectures will be held to prepare students for the midterm exam.\n- Midterm Exam Schedule: The exam will take place on Friday of next week, in-person, in the regular classroom.\n- Exam Rules and Requirements:\n - Students must bring a scientific calculator.\n - The usage of smartphones for calculations or as reference material is strictly prohibited.\n - The exam is closed notes.\n- Wednesday Discussion: Discussion sections will focus on reviewing for the midterm exam.\n- Exam Scope: Questions will be drawn from both the lectures and the discussion sections.\n- Supplemental Reading: One of the earliest RNA-seq papers (focused on bioinformatics and data analysis) has been uploaded; it will be covered in discussion sections after the midterm.\n\n# SOLiD Sequencing Technology Overview\n- Name: SOLiD stands for Support Oligonucleotide Ligation Detection, marketed by Applied Biosystems (ABI).\n- Market Position: While Illumina currently dominates the market, SOLiD was widely used until recently and its unique design principles remain valuable for future technological development.\n- Classification: Second-generation (or Next-Generation) sequencing technology.\n- Core Strategy: Uses a unique \u0022color space\u0022 strategy where sequencing is performed in colors which are subsequently converted into nucleotides.\n- Principle: Unlike Illumina's sequencing by synthesis, SOLiD uses sequencing by ligation.\n- Amplification: Utilizes emulsion PCR ($ePCR$) involving small beads ($DNA$ on beads) rather than bridge amplification.\n- Read Length: Typically limited to $75$ base pairs.\n\n# The SOLiD Sequencing Process\n## 1. Substrates and Emulsion PCR\n- The glass slide contains thousands of beads bound to the surface.\n- Each bead contains thousands of identical $DNA$ fragments amplified from a library ($cDNA$ or $DNA$).\n- Method: Emulsion PCR is conducted in oil-in-water droplets. Ideally, each droplet contains one bead with pieces of the template sequence amplified and immobilized to the bead via a universal adapter sequence called the $P1$ adapter.\n- Orientation: The $5'$ end of the template is immobilized to the bead, while the $3'$ end is free.\n\n## 2. Dye-Based Probes\n- Probe Structure: These are $8-mer$ oligonucleotides.\n- Orientation: Described from $3'$ to $5'$ direction.\n- Detailed Probe Composition:\n - First two nucleotides ($3'$ end): Dinucleotides representing all $16$ possible combinations ($4 \times 4$).\n - Color Labeling: There are only $4$ fluorescent colors. Each color corresponds to a specific subset of $4$ different dinucleotides. This mapping is predefined and fixed.\n - Next three nucleotides ($NNN$): Degenerate/random bases.\n - Final three nucleotides ($ZZZ$): Universal/fixed sequence.\n - Cleavage Site: Located between the $N$ and $Z$ positions.\n- Total Unique Sequences: There are $4^5 = 1024$ unique probe sequences ($4$ possible options for the two dinucleotide positions and the three random $N$ positions).\n\n## 3. Sequencing by Ligation Cycle\n- Primer Annealing: A universal sequencing primer ($n$) is annealed to the $P1$ adapter ($3'$ end of the template).\n- Ligation: All dye-based probes are introduced. $DNA$ ligase, which is sensitive to mismatches near the gap, ligates the probe that perfectly matches the $5$ template bases to the primer's phosphate group.\n- Imaging: An image is captured to identify the color of the probe, representing the specific dinucleotide category.\n- Capping: A phosphatase removes the phosphate group from any unextended strands to prevent further ligation, ensuring they do not create noise in later cycles.\n- Cleavage: A cleavage agent removes the $ZZZ$ portion and the fluorescent tag, leaving the $5$ bases ($dinucleotide + NNN$) ligated to the sequence.\n- Repetition: The cycle repeats. For example, $7$ cycles would result in a read length of $5 \times 7 = 35$ bases.\n- Primer Shifting: After a full round of sequencing, the strand is melted off, and a new primer ($n-1$) shifted by one nucleotide is used. This process is repeated five times with five different primers ($n$, $n-1$, $n-2$, $n-3$, and $n-4$).\n- Bridge Probes: These are cheaper, non-fluorescently labeled probes used to bridge the known adapter/primer regions so expensive reagents are not wasted on known sequences.\n\n# Color Space Decoding and Error Detection\n## Decoding Strategy\n- Double Interrogation: Because each base is interrogated twice (once in the first position of a dinucleotide and once in the second position due to primer shifting), each base is represented by two colors.\n- Decoding Logic: Decoding requires knowing the first base ($Base\,0$), which is the known last nucleotide of the adapter. If $Base\,0$ is known, the first color determines the identity of $Base\,1$. Then, $Base\,1$ and the next color determine $Base\,2$, and so on.\n\n## Distinguishing SNPs from Errors\n- Sequencing Error: Represents a wrong color call. Because of the chaining logic, a single color error would lead to extensive errors in converted base space. However, in color space, a sequencing error affects only $1$ color.\n- SNP: A real change in the nucleotide sequence ($Base\,Space$) affects $2$ adjacent colors because of the double interrogation.\n- Analysis Workflow: Reference genomes are converted into color space. Reads are aligned in color space, consensus is found in color space, and only then is the final consensus converted back to nucleotides.\n\n# Third-Generation Sequencing Technologies\n## Comparison to Second-Generation\n- Second-Gen (Illumina/SOLiD): Short reads (up to $150$), high coverage ($billions$), requires amplification, systematic errors, excellent for quantification.\n- Third-Gen (PacBio/Nanopore): Long reads ($kilobases$ to $megabases$), lower coverage ($10x$), no amplification (single molecule), random errors, excellent for genome assembly and structural analysis.\n\n## Pacific Biosciences (PacBio) - SMRT Sequencing\n- Name: Single Molecule Real Time ($SMRT$) DNA sequencing.\n- Hardware: Uses a \u0022SMRT cell\u0022 containing zero-mode waveguides ($ZMWs$), which are tiny pores approximately $70\,nm$ in diameter.\n- Mechanism: A DNA polymerase is immobilized at the bottom of each pore and captures a single template molecule. Gamma-labeled dNTPs (different colors for $A$, $C$, $G$, $T$) are added as the polymerase synthesizes a new strand.\n- Capture vs. Detection Volume: The camera focuses on the small capture volume at the bottom. Floating labeled nucleotides in the detection volume are the source of potential noise/error.\n- Circular Consensus Sequencing ($CCS$): Uses circular libraries with hairpin adapters. The polymerase sequences the same circle multiple times. Because errors are random, taking the consensus of these multiple passes dramatically reduces the error rate from $\approx 15\%$ to $0.1\%$.\n- Error Types: Primarily insertions and deletions ($indels$) due to stochastic timing and background signal.\n\n## Oxford Nanopore Sequencing\n- Mechanism: Uses protein-based nanopores embedded in a membrane. A motor protein pushes single-stranded $DNA$ through the pore while an electrical current is applied.\n- Detection: The sequencer measures changes in electrical current as different nucleotides pass through. These traces (\u0022squiggles\u0022) are interpreted using machine learning algorithms to call the bases.\n- Portability: Units like the MinION are the size of a USB drive and can be used in the field.\n- Direct RNA Sequencing: This is the only technology capable of sequencing RNA molecules directly without converting them to $cDNA$. This preserves RNA modification data (like methylation), which can be detected via specific current changes.\n- Limitations: Higher error rates (approx. $1\%$) compared to other methods, though constantly improving as new pore proteins are engineered.\n- Bias: Direct RNA sequencing can be limited by the stability of the RNA sample, as molecules may degrade during the process.