CSE 443 • Bioinformatics — Introductory Lecture
Course Logistics
Instructor
- Swakkhar Shatabda, Professor, Dept. of CSE, BRAC University
- Academic background: B.Sc. in CSE (BUET, ); Ph.D. (Griffith University, )
- Contact: swakkhar.shatabda@bracu.ac.bd · Office: 4G#28 · Counselling: Sun–Wed
- Research interests: Applied Machine Learning, Artificial Intelligence, Search, Computational Biology
Assessment Scheme
- Assignments (programming): — two tasks
- Quizzes: — best 3 of 4; no make-ups
- Mid-term exam:
- Final exam:
- Term project: (follow departmental policy for missed major exams)
Course Outline (12 Modules)
- 1. Intro to Bioinformatics
- 2. Sequence Alignment & Dynamic Programming
- 3. Hidden Markov Models (HMM), Profiles, Coding Regions
- 4. Genome Assembly: Read Mapping, Variant Calling, De Bruijn Graphs, Suffix Trees
- 5. Gene Expression Analysis: Clustering, Visualisation, Dimensionality Reduction
- 6. Single-Cell RNA-Seq & Spatial Transcriptomics
- 7. Phylogenetic Trees
- 8. Biological Networks
- 9. Gene Ontology & Biological Pathways
- Evolutionary & Comparative Genomics
- Representation & Contrastive Learning, Alignment in Bioinformatics
- Drug & Protein Structures, Proteomics, Generative AI
Practice Platform
- Rosalind (https://rosalind.info) — 284 textbook-track problems covering HMMs, peptide sequencing, etc.; live scoreboard motivates problem solving.
Reading Material
- B1: “Bioinformatics Algorithms: An Active Learning Approach” (Pevzner & Compeau, Vol I & II)
- B2: “Genome-Scale Algorithm Design” (Makinen et al.)
- B3: “Molecular Biology of the Cell” (Alberts et al.)
- B4: “A First Course in Systems Biology” (Voit)
- P1: Selected research papers
- H1: Technical handouts
Term-Project Workflow (max 4 members)
- Confirm group via Google Form, pick contemporary topic (link provided)
- Steps: understand problem ➜ literature review ➜ data experiments ➜ analysis & reporting
- Report template (Overleaf) shared; idea list on Google Docs
Foundational Definitions
- Bioinformatics (Merriam-Webster)
- Collection, classification, storage & analysis of biochemical/biological information using computers; especially molecular genetics & genomics.
- Bioinformatics (ChatGPT perspective)
- Interdisciplinary field combining biology + CS + math + statistics to interpret large-scale data (genomes, protein structures, expression profiles).
- Computational Biology
- Application of mathematical models & algorithms to understand/simulate biological systems; emphasises modelling, simulation & large-scale data analysis.
Human Diversity & Genotype ⇄ Phenotype
- Humans are identical at the DNA level; variation (≈ 1 in 1000 bases) drives diversity.
- A gene: segment of DNA encoding a specific protein → yields visible traits (eye/hair colour) & disease susceptibility.
- Typical gene contains nucleotide differences person-to-person, enough to alter:
- Protein shape/function, quantity, timing & location of expression.
- Genotype → Phenotype is modulated by environmental, epigenetic & stochastic factors; genome ≠ destiny.
Genome: Structure & Function
Genome = entire DNA content of an organism; inherited from each parent.
Physical facts: 6 billion base pairs in one human cell ≈ 6 ft linear length.
DNA packaged into chromosomes; genes may reside on either strand.
Upstream (5′) vs downstream (3′) nomenclature relative to gene start.
Example: Yeast genome snippets shown; illustrates continuous nucleotide strings, regulatory motifs & protein-coding genes embedded.
Genetic Diseases & Variants
- Mutation types
- Inherited (germline) — present at birth
- Acquired (somatic) — arise during lifetime (e.g.
cancer)
- Terminology: genetic variant = any DNA alteration; majority benign.
- Pathogenic effects: single-base indel or substitution can yield malfunctioning, excessive or absent protein ➜ disease; explains familial inheritance patterns.
DNA Chemistry & Topology
- Nucleotides
- Adenine (A) & Guanine (G): purines (two rings)
- Cytosine (C) & Thymine (T): pyrimidines (one ring)
- Base pairing
- A–T connected by H-bonds (weaker)
- C–G connected by H-bonds (stronger)
- Double Helix
- Complementary strands: if one strand reads …A C G…, opposite is …T G C…
- Directionality
- Phosphate backbone labels: 5′ → 3′; polymerases synthesise solely .
DNA Replication
- Watson–Crick insight: complementary pairing implies a copying mechanism.
- Steps
- Helicase unwinds; origin of replication opened.
- DNA polymerase reads template 3′ → 5′, adds nucleotides 5′ → 3′ on new strand.
- Leading strand: continuous synthesis; lagging strand: Okazaki fragments joined by ligase.
- Result: two duplexes (1 old + 1 new strand each).
RNA & Transcription
- Process
- RNA polymerase binds promoter/TSS, unwinds bubble, reads DNA 3′ → 5′, synthesises RNA 5′ → 3′.
- mRNA maturation (eukaryotes)
- 5′ cap (modified G)
- Splicing: remove introns, join exons
- 3′ poly(A) tail (chain of As)
- Key RNA Types
- mRNA: protein template
- tRNA: anticodon ↔ amino-acid adaptor
- rRNA: ribosome core
- snRNA: splicing machinery
- (others exist—miRNA, lncRNA, etc.)
- Comparing DNA vs RNA
- U instead of T
- Ribose (has 2′-OH) vs deoxyribose
- Single-stranded (often), more versatile structures
Translation & Genetic Code
- Codon = 3-nt mRNA word ⇒ maps to amino acid or stop.
- Total codes: ; 61 for amino acids, 3 as stop; AUG = start (Methionine).
- tRNA anticodon pairs with codon, delivers correct amino acid to ribosome ➜ peptide elongates N-terminus → C-terminus.
Proteins
- Polymer of 20 amino acids; primary sequence dictates folding & function.
- Functions: enzymes, structural (contractile), transporters, immune (antibodies), signalling, receptors, etc.
- Higher-order structure (secondary, tertiary, quaternary) arises from interactions (H-bonds, hydrophobic packing, disulfide, ionic).
Gene Regulation
- Cells don’t express all genes simultaneously.
- Promoter region (upstream 5′): contains DNA motifs recognised by transcription factors (TFs); binding recruits/blocks RNA polymerase ⇒ modulates transcription rate.
- Regulatory network integrates signals (environmental, developmental) to achieve cell-type-specific expression (explains neuron vs skin cell from same genome).
Key Philosophical Points
- “Same code, different outcome” — epigenetic & regulatory layers give rise to diverse cell types.
- Genome provides blueprint, but phenotype emerges from multilayered control + environment.
Next-Class Teaser
- Question posed: “Where does DNA start its replication?”
- Students to fill Google Form link; will explore origins of replication (OriC, etc.) in subsequent lecture.