Computational Biology

(BIOSC 1540)

Feb 11, 2025

Lecture 06A

Read mapping

Foundations

Let's have a discussion

Let's have a discussion

Changes I can make

A. Simpler and fewer Python problems

B. Offer an optional Python recitation either in the evening or weekend

C. "Flipped" classroom where I record lectures/assign readings and use classtime for Python

D. No changes

E. No Python

Announcements

After today, you should have a better understanding of

How transcriptomics extends beyond genomics

DNA

Genomics allows us to study the blueprint of organisms

DNA sequences are stable across an organism’s lifetime

Genomics helps answer key biological questions

Genomics tells us what’s possible for an organism to do but not when or how it does it.

A genomic sequence alone doesn’t tell us what genes are active

After today, you should have a better understanding of

How transcriptomics extends beyond genomics

RNA

Transcriptomics: A real-time microscope

• Environmental conditions: Cells respond to stress, nutrients, or pathogens by changing gene expression
• Developmental stage: The genes active in an embryo differ from those in an adult
• Cell type: A neuron will have a different gene expression profile than a liver cell

Genomics Provides a Static Blueprint, but Transcriptomics Captures Dynamic Activity

Transcriptomics works with the complete set of RNA transcripts

Transcriptomics reveals alternative splicing and isoforms

Example: Dscam in Drosophila

Functional insights

• Captures real-time cellular responses

After today, you should have a better understanding of

The role of RNA-seq in modern transcriptomics

Sample collection

Separate cells from media

Great! We have our cells, but how can we extract our RNA?

The first step is always to centrifuge and separate our cells and media

Keep the part that has our component of interest (RNA)

We break open our cells by lysing them

Chemical lysis destabilizes the lipid bilayer and denatures proteins

Surfactants have a hydrophilic head and hydrophobic tail

Phenol-chloroform extraction exploits solubility and density differences

Reverse transcription introduces unique challenges

RNA is converted to cDNA using reverse transcriptase

• Random or oligo(dT) primers influence transcript representation
• Second-strand synthesis method can preserve strand information

mRNA enrichment focuses sequencing on protein-coding transcripts

After today, you should have a better understanding of

The role of RNA-seq in modern transcriptomics

Sample quality

RNase starts degrading RNA rapidly

RNA quality is critical for successful sequencing

• rRNA makes up a large (~85%) of our RNA
• Based on the ratio of 28S and 18S rRNA vs. all RNA

After today, you should have a better understanding of

The role of RNA-seq in modern transcriptomics

Microarrays

Once upon a time, we had microarrays

Microarrays have some caveats

• Limited to known sequences: Can only detect pre-defined sequences
• Cross-hybridization: Similar sequences may cause false positives
• Limited dynamic range: May miss very low or high abundance transcripts
• Normalization challenges: Complex process, potential for bias

After today, you should have a better understanding of

The role of RNA-seq in modern transcriptomics

RNA-seq

RNA sequencing changed the game

Now we just sequence the cDNA

• RNA-seq doesn't require prior knowledge of sequences
• Enables discovery of novel transcripts and isoforms
• Provides absolute quantification rather than relative concentration

Advantages

TopHat questions

What is the primary advantage of RNA-seq over microarray technology?

After today, you should have a better understanding of

Why read mapping is essential for transcriptomics

Read Mapping is Essential for Making Sense of RNA-seq Data

• RNA-seq produces millions of short sequencing reads (~50–150 bp).
• These reads must be correctly mapped to a reference genome or transcriptome to identify which genes are being expressed.
• Proper alignment enables:
  • Gene expression quantification – Counting mapped reads per gene.
  • Isoform detection – Distinguishing different transcript variants.
  • Splicing analysis – Identifying exon-exon junctions.

The Goal of Read Alignment is to Reconstruct Gene Expression Patterns

• Each RNA-seq read represents a small fragment of a transcript.
• By mapping reads to a reference genome or transcriptome, we can:
  • Identify which genes are active in a sample.
  • Measure the relative abundance of different transcripts.
  • Detect novel isoforms and alternative splicing events.
• Without accurate alignment, downstream analysis (e.g., differential expression) is unreliable.

Challenges in Aligning Short Reads to a Large Reference Genome

The human genome is ~3 billion bases, but RNA-seq reads are only ~100 bases long.

A naïve approach would require searching for every read across billions of bases, which is computationally infeasible.

Why is this a problem?

• Reads may match multiple locations (repetitive sequences).
• Sequencing errors create mismatches, making alignment difficult.
• Polymorphisms & mutations in different individuals affect alignment accuracy.

Why Transcriptomic Read Mapping is Different from Genomic Read Mapping

• Unlike DNA sequencing, RNA sequencing includes spliced transcripts.
• Key problem: Reads from mRNA span exon-exon junctions, but the genome contains introns.
  • Example: A read from an mRNA transcript might originate from exon 1 and exon 2, but the genome has a large intron in between.
  • Consequence: Traditional genomic aligners fail to align these reads correctly.
• Solution: Transcriptomic aligners must allow for gapped alignments that bridge exon-exon junctions.

Splice Junctions Complicate Read Mapping

• Genes contain introns that are removed during splicing, but these are not present in mature RNA.
• Example: A 100-bp RNA-seq read may contain 50 bp from Exon 1 and 50 bp from Exon 2, skipping a 10,000-bp intron.
• Challenges:
  • Reads spanning junctions do not exist as contiguous sequences in the genome.
  • Aligners must infer intron-exon boundaries from known or novel splice sites.
• Specialized RNA-seq aligners (e.g., STAR, HISAT2) are designed to handle this.

Isoforms Add Another Layer of Complexity to Read Mapping