B.Tech. / M.Tech. (Integrated) DEGREE EXAMINATION, JULY 2023
21BTB102T - INTRODUCTION TO COMPUTATIONAL BIOLOGY

Detailed Solution:
Step 1: Biological classification aims to group organisms based on evolutionary relationships.
Step 2: In 1969, the American plant ecologist R.H. Whittaker proposed the widely accepted Five Kingdom classification system.
Step 3: These five kingdoms are Monera, Protista, Fungi, Plantae, and Animalia, classified based on cell structure, body organization, mode of nutrition, and reproduction.

Detailed Solution:
Step 1: Ribosomes are non-membrane-bound organelles found in both prokaryotes (70S) and eukaryotes (80S).
Step 2: Regardless of the organism type, the sole biological function of a ribosome is translation.
Step 3: Translation is the process of synthesizing proteins by linking amino acids together in an order specified by messenger RNA (mRNA).

Detailed Solution:
Step 1: The complete oxidation of glucose (aerobic cellular respiration) involves Glycolysis, the Krebs cycle, and the Electron Transport Chain.
Step 2: While glycolysis happens in the cytoplasm, the major energy-yielding steps (Krebs cycle and Oxidative Phosphorylation) occur inside a specific double-membraned organelle.
Step 3: This organelle is the Mitochondria, often referred to as the powerhouse of the cell because it produces the vast majority of the cell's ATP (energy).

Detailed Solution:
Step 1: In Prophase, chromosomes condense. In Metaphase, they align at the cell equator.
Step 2: In Anaphase, the enzyme separase cleaves the cohesin proteins holding the sister chromatids together at the centromere.
Step 3: The spindle fibers then rapidly retract, pulling the separated sister chromatids (now considered individual chromosomes) to opposite poles of the cell.

Detailed Solution:
Step 1: In all living cellular organisms (bacteria, archaea, and eukaryotes), the primary molecule used for long-term storage of hereditary instructions is Deoxyribonucleic Acid.
Step 2: RNA generally serves as a temporary messenger (mRNA) or structural/catalytic component (rRNA, tRNA) to help express the genetic information, but the stable master archive is DNA.

Detailed Solution:
Step 1: Polysaccharides are long polymeric chains consisting of hundreds or thousands of monosaccharide units (e.g., Starch, Cellulose, and Glycogen are all polymers of glucose).
Step 2: Lactose (milk sugar) is a disaccharide, meaning it consists of exactly two monosaccharides joined together: one glucose and one galactose molecule.

Detailed Solution:
Step 1: DNA uses four standard nitrogenous bases: Adenine (A), Thymine (T), Cytosine (C), and Guanine (G).
Step 2: RNA uses A, C, and G, but replaces Thymine with Uracil (U).
Step 3: Therefore, Uracil is the base that is exclusively found in RNA and is NOT present in DNA.

Detailed Solution:
Step 1: Let's evaluate the options. PDB (Protein Data Bank) stores 3D structures. UniProtKB stores protein sequences and functions. PROSITE stores protein domains and families.
Step 2: GenBank, maintained by the NCBI, is the NIH genetic sequence database. It is a comprehensive, open-access, annotated collection of all publicly available nucleotide (nucleic acid) sequences and their protein translations.

Detailed Solution:
Step 1: During translation, the ribosome scans the mRNA until it finds the specific signal to begin protein synthesis.
Step 2: This initiation signal is the start codon, which is universally AUG. It codes for the amino acid Methionine (or formyl-methionine in prokaryotes).
Step 3: UAA, UAG, and UGA are the three stop codons that terminate translation.

Detailed Solution:
Step 1: Proteins are polymers of amino acids.
Step 2: These amino acids are linked together end-to-end to form a primary polypeptide chain.
Step 3: While the bond is technically a type of covalent amide bond, its highly specific biological name in the context of proteins is the Peptide bond. (Glycosidic bonds are found in carbohydrates).

Detailed Solution:
Step 1: Insulin is a hormone regulating glucose. Trypsin is a digestive enzyme. Testosterone is a steroid hormone.
Step 2: Hemoglobin is an iron-containing metalloprotein found inside red blood cells.
Step 3: Its primary biological role is to reversibly bind to oxygen in the lungs and transport it through the bloodstream to release it in the body's tissues.

Detailed Solution:
Step 1: Hair, nails, hooves, horns, and feathers are incredibly tough, structural biological materials.
Step 2: They derive their mechanical strength and insolubility from a specific family of fibrous structural proteins.
Step 3: This protein family is called Keratin. (Actin is for muscle contraction, Albumin is a blood plasma protein).

Detailed Solution:
Step 1: The nervous system contains neurons and supporting glial cells.
Step 2: The most abundant type of macroglial cell in the Central Nervous System has numerous branching processes extending from the soma.
Step 3: This morphology resembles a star, hence they are aptly named Astrocytes (from Greek astron = star, kyto = cell).

Detailed Solution:
Step 1: An electrical signal (action potential) travels down the axon of a neuron but cannot jump the physical gap (synapse) to the next neuron.
Step 2: To bridge this gap, the electrical signal is converted into a chemical signal.
Step 3: The presynaptic neuron releases chemical messengers called Neurotransmitters, which diffuse across the synapse and bind to receptors on the next neuron, triggering a new electrical signal.

Detailed Solution:
Step 1: Parkinson's is caused by loss of dopamine neurons. Multiple sclerosis by demyelination. Spongiform encephalopathy (prion diseases) by PrPSc plaques.
Step 2: The defining histological hallmark of Alzheimer's disease is the extracellular accumulation of amyloid-beta ($Aeta$) proteins in the brain.
Step 3: These toxic clumps are specifically referred to as amyloid plaques, which disrupt cell-to-cell communication and trigger inflammation.

Detailed Solution:
Step 1: Artificial Neural Networks (ANNs) mathematically mimic biological neurons.
Step 2: In a biological neuron, the Dendrites are the branch-like structures that receive incoming signals from other cells.
Step 3: Therefore, the input units ($x_1, x_2, ... x_n$) of an artificial neuron directly correspond to the signal-receiving dendrites of a physical neuron. (The Axon represents the output).

Detailed Solution:
Step 1: All blood cells, including all lymphocytes, originate from hematopoietic stem cells.
Step 2: These stem cells are located in the red Bone marrow.
Step 3: While T-lymphocytes migrate to the Thymus to mature, B-lymphocytes both originate and fully mature within the Bone marrow (which is why they are called "B" cells, originally discovered in the Bursa of Fabricius in birds, but analogous to Bone marrow in humans).

Detailed Solution:
Step 1: Live attenuated and inactivated vaccines use the whole intact pathogen (either weakened or dead).
Step 2: A vaccine that is purified to include only the specific parts of the microbe that best stimulate the immune system (like a specific surface protein or polysaccharide cap) is called a Subunit vaccine.
Step 3: Because they only contain the essential antigens and not the whole organism, subunit vaccines are extremely safe and cannot cause the disease (e.g., the Hepatitis B or HPV vaccines).

Detailed Solution:
Step 1: Allergic reactions (Type I hypersensitivity) are driven by the sudden release of inflammatory mediators like histamine.
Step 2: The primary cells responsible for storing and releasing histamine are tissue Mast cells and circulating Basophils.
Step 3: When an allergen binds to IgE antibodies on the surface of a basophil, it triggers rapid degranulation, releasing the chemicals that cause the classic symptoms of an allergic reaction (vasodilation, itching, swelling).

Detailed Solution:
Step 1: Unlike B-cells which can recognize free-floating 3D antigens, T-cells are strictly "MHC-restricted."
Step 2: A T-cell receptor (TCR) can only recognize a short linear peptide epitope if it is securely bound and presented on the surface groove of a Major Histocompatibility Complex (MHC) molecule.
Step 3: Therefore, the entire foundation of computational T-cell epitope prediction relies on predicting the binding affinity between a peptide fragment and specific MHC alleles. (While "peptide binding" is the action, MHC is the critical biological context that defines a T-cell epitope).

1. Introduction to Prokaryotic Cells:
Prokaryotes (from Greek pro = before, karyon = nucleus) are the simplest, oldest, and most abundant cellular organisms on Earth, primarily encompassing Bacteria and Archaea. They are unicellular organisms characterized by the complete absence of a membrane-bound true nucleus and other membrane-bound organelles (like mitochondria or ER). They are typically small (0.1 to 5.0 μm).

2. Structure and Function of Prokaryotic "Organelles" and Components:

• Nucleoid Region: An irregularly shaped region in the cytoplasm where the cell's single, circular, double-stranded DNA chromosome is concentrated. Function: Stores the primary genetic information essential for survival and reproduction.
• Plasmids: Small, extra-chromosomal circular DNA molecules. Function: Carry accessory genes (like antibiotic resistance) that can be swapped with other bacteria via conjugation.
• Ribosomes (70S): Molecular machines made of RNA and protein floating free in the cytoplasm. Function: The exclusive sites of protein synthesis (translation).
• Plasma Membrane: A phospholipid bilayer surrounding the cytoplasm. Function: Controls the passage of substances in and out of the cell. In prokaryotes, it also houses the enzymes required for cellular respiration (since they lack mitochondria).
• Cell Wall: A rigid outer layer outside the plasma membrane (made of peptidoglycan in bacteria). Function: Provides structural shape and protects against osmotic bursting in hypotonic environments.
• Capsule: A sticky, jelly-like outer coating present in some bacteria. Function: Protects the cell from dehydration and helps it evade the host's immune system (phagocytosis).
• Flagella: Long, whip-like tail structures. Function: Provide motility, acting like a propeller to move the bacterium through fluids.
• Pili / Fimbriae: Hair-like appendages on the surface. Function: Fimbriae help attach to surfaces (like host tissues), while specialized sex pili allow DNA transfer between bacteria.

PART A: Enzymes
1. Description: Enzymes are biological catalysts, almost exclusively made of proteins (with specific 3D shapes). They accelerate the rate of chemical reactions in the body by lowering the activation energy required, without being consumed or permanently altered in the process. They are highly specific to their substrates (lock and key mechanism).
2. Functions: They drive almost every biochemical pathway, including digestion, DNA replication, cellular respiration, and detoxification.
3. Examples:

• Amylase: Found in saliva; breaks down complex starch (carbohydrates) into simple sugars.
• DNA Polymerase: Synthesizes new DNA strands during cell division.
• Lactase: Breaks down lactose (milk sugar) in the small intestine.

PART B: Hormones
1. Description: Hormones are chemical messengers secreted by the ductless glands of the endocrine system directly into the bloodstream. They can be proteins, peptides, or steroids. Unlike enzymes that act locally and quickly, hormones travel long distances to target distant organs, inducing slow, long-lasting physiological changes.
2. Functions: They regulate homeostasis, growth, development, metabolism, reproduction, and mood.
3. Examples:

• Insulin: Secreted by the pancreas; lowers blood glucose levels by signaling cells to absorb glucose.
• Thyroxine (T4): Secreted by the thyroid gland; regulates the body's basal metabolic rate.
• Adrenaline (Epinephrine): Secreted by the adrenal glands; triggers the rapid "fight or flight" response by increasing heart rate and blood flow to muscles.

1. Introduction to Protein Visualization:
Proteins are complex 3D macromolecules whose function is entirely dependent on their shape. The Protein Data Bank (PDB) stores the XYZ atomic coordinates of these structures. Protein visualization tools (molecular graphics software) are computer programs that render these numerical coordinates into interactive 3D graphical representations (like ribbons, spheres, or surfaces), allowing scientists to study binding sites, mutations, and structural architecture.

2. Key Protein Visualization Tools:

• PyMOL:
  Description: One of the most popular, open-source molecular visualization systems. Written in Python, it is highly customizable via scripting.
  Application: It is the industry standard for creating extremely high-quality, publication-ready 3D images and animations of proteins. It excels at rendering complex protein-ligand interactions and surface electrostatics.


• UCSF Chimera (and ChimeraX):
  Description: Developed by the University of California, San Francisco, it is a highly extensible program for interactive visualization and analysis.
  Application: Best known for its ability to handle immense datasets, particularly visualizing massive structures like virus capsids or integrating structural data with Cryo-Electron Microscopy (Cryo-EM) density maps.


• RasMol:
  Description: One of the earliest and most historically significant visualization tools. It is lightweight and command-line driven.
  Application: While largely superseded by PyMOL, it remains a fantastic educational tool for quickly viewing basic representations (wireframe, space-filling, ribbon) of a PDB file on low-end hardware.


• Jmol / JSmol:
  Description: An open-source Java/JavaScript viewer for chemical structures.
  Application: Because it runs entirely within a web browser without requiring software installation, it is universally used on biological databases (like the RCSB PDB website itself) to provide instant, interactive 3D previews of proteins to users online.

1. What are Artificial Neural Networks (ANNs)?
An Artificial Neural Network is a computational framework and machine learning algorithm inspired by the architecture of the biological human brain.
- It consists of interconnected artificial "neurons" (nodes) arranged in layers: an Input layer (receiving data), one or more hidden Middle layers (processing data), and an Output layer (delivering the prediction).
- Connections between nodes have mathematical "weights." As the network is fed training data, it iteratively adjusts these weights (via backpropagation) to "learn" patterns, enabling it to classify new, unseen data or make highly accurate predictions without explicit programming.

2. Applications of Neural Networks in Biology (Bioinformatics):
Because biological data (like genomics and proteomics) is massive, noisy, and non-linear, ANNs are the perfect tool to analyze it.

• Protein Structure Prediction: The most famous biological application is DeepMind's AlphaFold, a deep neural network that solved the 50-year-old protein folding problem by predicting 3D tertiary structures from 1D amino acid sequences with unprecedented accuracy.
• Genomics & Sequence Analysis: ANNs are used to scan entire DNA sequences to identify hidden patterns, such as finding splice sites, predicting gene locations (promoters and exons), and classifying sequences as coding or non-coding.
• Medical Imaging & Diagnostics: Convolutional Neural Networks (CNNs) are trained on thousands of medical images to detect anomalies. They can analyze histological tissue slides to diagnose cancer (e.g., melanoma from skin images) faster than human pathologists.
• Drug Discovery: Neural networks can predict how effectively a newly designed chemical drug will bind to a biological target (like a viral enzyme) and predict its potential toxicity, saving years of physical lab testing.
• Immunology: ANNs are heavily used in reverse vaccinology to predict which specific peptide sequences (epitopes) of a virus will successfully bind to human MHC molecules and trigger an immune response.

1. Introduction to Vaccines:
A vaccine is a biological preparation that provides active acquired immunity to a particular infectious disease. It works by introducing a safe, modified version (or a part) of a pathogen into the body. This safely triggers the immune system to produce antibodies and memory B/T cells, preparing the body to mount an immediate defense if it ever encounters the actual, live pathogen.

2. Main Types of Vaccines:

• Live Attenuated Vaccines:
  Mechanism: Contains a live version of the virus or bacteria that has been severely weakened (attenuated) in the lab so it cannot cause disease in healthy people.
  Pros/Cons: Elicits the strongest, longest-lasting cellular and humoral immune response (often lifelong with one dose). However, it is unsafe for immunocompromised individuals.
  Examples: Measles, Mumps, Rubella (MMR), Oral Polio Vaccine (Sabin).


• Inactivated (Killed) Vaccines:
  Mechanism: The pathogen is grown in a lab and completely killed/destroyed using heat, chemicals (like formaldehyde), or radiation.
  Pros/Cons: Extremely safe (cannot revert to a virulent form). However, the immune response is weaker than live vaccines, often requiring multiple "booster" shots over time.
  Examples: Rabies vaccine, Injected Polio Vaccine (Salk).


• Subunit, Recombinant, and Conjugate Vaccines:
  Mechanism: Instead of the whole microbe, these use only the specific, purified pieces of the pathogen that best stimulate the immune system (like a specific surface protein or sugar capsule).
  Pros/Cons: Very safe with very few side effects. Good for people with weakened immune systems.
  Examples: Hepatitis B, HPV, Pneumococcal conjugate vaccine.


• Toxoid Vaccines:
  Mechanism: Some bacterial diseases are caused by the toxins the bacteria secrete, rather than the bacteria themselves. Toxoid vaccines contain deactivated toxins (toxoids) that train the immune system to fight the toxin.
  Examples: Tetanus, Diphtheria.


• mRNA (Nucleic Acid) Vaccines:
  Mechanism: A modern breakthrough. They do not contain any part of the pathogen. Instead, they deliver a strip of messenger RNA encased in a lipid nanoparticle. The mRNA instructs host cells to temporarily manufacture a harmless piece of the viral protein (e.g., spike protein), triggering an immune response.
  Examples: Pfizer and Moderna COVID-19 vaccines.

1. What are Stem Cells?
Stem cells are unspecialized "master" cells of the body. They are defined by two unique properties: Self-renewal (the ability to divide endlessly while remaining unspecialized) and Potency (the ability to differentiate and mature into specialized functional cells, like muscle, brain, or blood cells).

2. Classification of Stem Cells based on Potency:

• Totipotent: Can differentiate into any cell type in the body, PLUS extra-embryonic tissues (the placenta). E.g., the Zygote and cells of the very early embryo (up to 8-cell stage).
• Pluripotent: Can differentiate into any of the >200 cell types that make up the human body (derived from the three germ layers: endoderm, mesoderm, ectoderm), but cannot form an entire organism because they cannot make a placenta. E.g., Embryonic Stem Cells (ESCs) taken from the inner cell mass of a blastocyst.
• Multipotent: Can only differentiate into a closely related family of cells within a specific tissue. E.g., Adult Hematopoietic (blood) stem cells in the bone marrow can become red blood cells, white blood cells, or platelets, but cannot become neurons.

3. Applications of Stem Cells:

• Regenerative Medicine (Cell Therapy): The most profound application is replacing diseased or damaged tissues. Bone marrow transplants (using hematopoietic stem cells) are routinely used to cure blood cancers like leukemia. In the future, stem cells could be used to regenerate insulin-producing cells for Type-1 diabetics, or cardiac muscle after a heart attack.
• Neurological Treatments: Stem cells hold immense promise for replacing damaged neurons to treat currently incurable conditions like Parkinson's disease, Alzheimer's, or severe spinal cord injuries.
• Drug Testing: Pharmaceutical companies can grow human stem cells into specific tissues (like human liver cells) in a dish. This allows for safe, accurate testing of new drugs for efficacy and toxicity without risking human patients.
• Disease Modeling: Using Induced Pluripotent Stem Cells (iPSCs—adult cells reprogrammed back to an embryonic state), scientists can grow cells from a patient with a genetic disorder to study the exact cellular mechanics of the disease in the lab.

PART A: Biological Databases
1. Introduction:
A biological database is a large, organized collection of biological data (such as DNA sequences, protein sequences, 3D structures, and metabolic pathways) stored on computer servers. They are the foundational infrastructure of modern bioinformatics, allowing scientists globally to store, query, and analyze biological information.

2. Primary Categories of Databases:

• Nucleotide Sequence Databases: Store raw DNA/RNA sequences. The major three (which synchronize data daily via the INSDC) are:
  • GenBank: Maintained by NCBI (USA).
  • EMBL-Bank: Maintained by the European Bioinformatics Institute (Europe).
  • DDBJ: DNA Data Bank of Japan.
• Protein Sequence Databases: Store amino acid sequences.
  • UniProtKB: The central hub for protein data, consisting of Swiss-Prot (manually curated, high quality) and TrEMBL (computationally translated).
• Structural Databases:
  • PDB (Protein Data Bank): The worldwide archive containing the 3D XYZ atomic coordinates of experimentally solved biological macromolecules (proteins, DNA, RNA).

---

PART B: The BLAST Algorithm for Sequence Search
1. What is BLAST?
BLAST (Basic Local Alignment Search Tool) is the most widely used bioinformatics algorithm. When a biologist sequences an unknown piece of DNA or protein, they use BLAST to compare it against a massive database (like GenBank) to find regions of local similarity, which helps infer the evolutionary origin and function of the unknown sequence.

2. Why use BLAST instead of Exact Matching?
Due to evolution and mutations (insertions, deletions, substitutions), a sequence is rarely a 100% exact match. Exact algorithms (like Needleman-Wunsch) are mathematically perfect but too slow for databases with billions of letters. BLAST is a heuristic algorithm—it trades mathematical perfection for incredible speed, finding highly probable matches in seconds.

3. How the BLAST Algorithm Works (The 3 Phases):

• Phase 1: Seeding (Creating Words)
  BLAST breaks the user's query sequence into short, manageable "words" (default length is 3 for proteins, 11 for DNA). For example, a protein query 'PQGE' yields words 'PQG' and 'QGE'. It then creates a list of synonymous words that score highly based on a substitution matrix (like BLOSUM62).
• Phase 2: Scanning (Finding Hits)
  BLAST rapidly scans the massive target database looking only for exact matches to this list of short words. Because it only looks for short exact matches, this step is extremely fast. When it finds a match, it is called a "hit."
• Phase 3: Extension (Local Alignment)
  Once a hit is found, BLAST anchors on that spot and tries to extend the alignment in both directions (left and right) along the sequence. As it extends, it calculates a score (adding points for matches, subtracting for mismatches/gaps). It continues extending until the score drops below a certain threshold. The resulting high-scoring segment pairs (HSPs) are reported to the user as local alignments.

4. Output Evaluation (E-value):
BLAST provides an E-value (Expect value) for every match. The E-value indicates the number of hits one can "expect" to see by random chance. An E-value close to zero (e.g., $1e-50$) indicates a highly significant evolutionary match that is definitively not due to random chance.

1. Introduction: The Biological Data Explosion
With the advent of High-Throughput Technologies (Next-Generation DNA Sequencing, Mass Spectrometry, Microarrays), biology transitioned from a data-poor to a data-rich science. The databases are now expanding at an exponential rate (Petabytes of data). Traditional manual analysis is impossible. Data Mining and Machine Learning (ML) are the computer science disciplines utilized to automatically sift through this "Big Data" to extract meaningful biological knowledge.

2. Data Mining in Biology:
Data mining is the process of discovering hidden patterns, correlations, and anomalies within large datasets.

• Sequence Mining: Scanning massive DNA databases to find recurrent motifs (like TATA boxes or Transcription Factor Binding Sites) that indicate regulatory regions of genes.
• Microarray Data Analysis (Clustering): Gene expression arrays produce thousands of data points. Data mining tools use clustering algorithms (like K-means or Hierarchical clustering) to group genes that express similarly under certain conditions. If a group of unknown genes clusters with known cancer-causing genes, their function can be inferred.
• Text Mining: Millions of biological research papers exist on PubMed. Natural Language Processing (NLP) algorithms mine this text to automatically build networks of interacting proteins and genes without a human reading a single paper.

3. Machine Learning in Biology:
Machine Learning is the mechanism powering advanced data mining. It involves algorithms that "learn" from data and improve their performance over time without explicit programming.

A. Supervised Learning:
The algorithm is trained on labeled data (e.g., a dataset of images labeled "cancerous" vs "healthy").

• Medical Diagnostics: Support Vector Machines (SVMs) or Convolutional Neural Networks (CNNs) are trained on thousands of MRI scans. Once trained, they can classify new, unseen MRIs to detect early-stage tumors with precision equal to or exceeding human doctors.
• Gene Finding: Hidden Markov Models (HMMs) are trained on known gene sequences. They then scan newly sequenced genomes to predict where the new genes are located.

B. Unsupervised Learning:
The algorithm is given unlabelled data and must find the inherent structure itself.

• Disease Subtyping: By feeding the genetic profiles of thousands of breast cancer patients into a clustering algorithm, ML can discover that what we thought was one disease is actually four genetically distinct subtypes of breast cancer, each requiring a different targeted therapy.

C. Deep Learning (The Modern Era):
Utilizing deep Artificial Neural Networks with many layers, Deep Learning has revolutionized biology.

• Protein Folding (AlphaFold): Google DeepMind used deep learning to predict the 3D tertiary structure of proteins directly from their 1D amino acid sequence, effectively solving a 50-year-old grand challenge in biology.
• Drug Discovery: ML models simulate and predict how millions of chemical compounds will dock with a viral protein, radically speeding up the discovery of new drugs (like predicting COVID-19 antiviral candidates in days instead of years).

4. Conclusion:
Data mining and Machine learning have transformed biology from a purely experimental science into a predictive and computational science, enabling the modern era of personalized medicine, rapid drug design, and deep systemic understanding of life.