Genome Browsers Overview Quiz

Genome Browsers

• UCSC, NCBI, and Ensembl are the “Big 3” genome browsers
• NCBI includes ClinVar, a database with publicly submitted variations, and dbSNP, a database of SNPs
• GnomAD is a genome aggregation database and also a good source for variant information

UCSC

• UCSC Genome Browser is available at http://genome.ucsc.edu/
• UCSC provides the Track View and Table Browser
• UCSC offers two assemblies: GRCh38 and GRCh37 with different levels of annotation
• GRCh38 has fewer gaps and is newer, while GRCh37 is older and has more annotations

NCBI

• NCBI is available at https://www.ncbi.nlm.nih.gov/
• NCBI focuses on database queries.
• ClinVar displays submitted variations and observed disease associations
• dbSNP (a database of SNPs) is also housed within NCBI
• GEO (Gene Expression Omnibus) offers over 100,000 expression experiments
• PubMed lists publications

Ensembl

• Located at http://useast.ensembl.org/index.html
• Ensembl is a European Genome Browser
• Ensembl combines features of UCSC and NCBI, such as Track Viewing and Table Browser

Gnomad

• Gnomad is available at https://gnomad.broadinstitute.org/
• Provides information on position, change, minor allele frequency (MAF), consequence, and clinical significance
• Gnomad is frequently updated, so the content in screenshots may not always be the same.

UCSC Anatomy

• Coordinates can be entered by gene symbol or coordinates
• Track View displays various data types
• You can zoom in/out and pan left/right
• You can access DNA view and analyze at the amino acid level
• The strand direction reveals if the gene is on the reverse (3’ to 5’) strand

UCSC Tracks

• The Track View is organized by type, such as mapping, genes, phenotype, etc.
• Users can turn on/off different tracks of data
• Tracks provide links to additional information

NCBI

• NCBI's data extraction tools are complicated

Ensembl Automation

• Ensembl provides automated extraction tools
• It offers a Perl API for programmatic data extraction and a REST (representational state transfer) interface that supports Perl, Python, Java, and other languages

Proteomics

• Proteins need to be experimentally verified to exist and for their properties to be measured
• Post-translational modifications change protein behavior from what the genetic sequence alone indicates
• Addition of a phosphate by a kinase is a post-translational modification
• Addition of sugar to a protein is a post-translational modification, like in the COVID-19 spike protein

Vitamin A Deficiency

• 250,000 to 500,000 children become blind every year due to vitamin A deficiency
• Data on vitamin A deficiency worldwide collected from 1991 to 2013

Golden Rice

• Genetically modified to contain beta-carotene, a precursor to vitamin A
• Anti-GMO groups fought against the introduction of golden rice in Bangladesh

Concerns about GMOs

• A 2017 review found no evidence that GMO animal feed has adverse effects on animal health
• Trillions of animals have been fed with GMO feed without negative health consequences
• Switching to non-GMO feed has negative consequences
  • Increases greenhouse gas emissions by 7%
  • Increases land use
  • Raises food prices

Mutation Breeding

• Modern agriculture has practiced mutation breeding for the last 100 years
• Chemicals or radiation are used to increase mutations, resulting in new varieties
• This process is not considered GMO
• A 2017 PNAS report found no evidence that foods from GE crops are less safe than foods from non-GE crops

Earth BioGenome Project

• Proposed to sequence the genomes of all named extant eukaryotes, about 2 million species
• Aims to create a digital library of life on earth
• Projected to cost $4.7 billion

Metagenomics

• Analysis of all the genomics in a coherent environmental sample
• Examples of environments studied: 1 ml of ocean water, 1 gram of soil, and human gut

Applications of Metagenomics

• Studies human gut biome and its relation to health
• Applies to agriculture, environmental remediation, anthropology, and biotech

Gene Terminology

• Paralogues are genes that diverged from a common ancestor
• Orthologues are genes that perform the same function in different species

Pseudogenes

• Are typically non-functional genes
• Appear like existing gene sequences
• Likely arise from DNA duplication or retro-transposition

Pseudogene.org

• Comprehensive database and comparison platform for pseudogene annotation
• Estimates approximately 10,668 processed pseudogenes and 14,000 non-processed pseudogenes

PTEN: Phosphatase and Tensin Homolog

• Is a phosphatase, which removes phosphates from a protein

Gene Names

• Ambiguous, have aliases, synonyms, and historical names
• Example: Hemoglobin Subunit Gamma 1

Globin Gene Clusters

• Located on chromosomes 16 and 11

Human Genome

• Contains 3.2 * 109 nucleotides (haploid)
• Chromosome sizes range from 49 Mbp to 279 Mbp
• Repeated sequences make up 50% of the genome

Cell Senescence

• Can result from the normal shortening of telomeric DNA
• Cells irreversibly withdraw from the cell cycle and fail to respond to proliferation-inducing stimuli

RNA

• Thousands of transcripts (alternatively spliced) in a human genome
• LncRNAs: Long-non-coding RNAs (length > 200 nucleotides)
• MicroRNAs: ~22 nucleotides in length whose function is RNA silencing

SNPs

• Single Nucleotide Polymorphisms
• Compare any 2 un-related individuals, they will be different at approximately 0.1% of sites (1 in 1,000).
• 3109 * 0.001 = 3107 = 30,000,000 differences

Privacy Concerns

• 20 million people have uploaded their genetic profiles to consumer DNA sites
• Concerns that police can access this genetic data to solve crimes
• First case in which a judge approved a warrant to penetrate GEDmatch and search its database

Case Example

• A Florida jury found Thomas Garner guilty of first-degree murder for the 1984 death of Navy recruit Pamela Cahanes using genetic data

Atomic Level Interactions

• DNA binding is a significant process at the atomic level.

Cross-Species Genome Similarity

• "Islands of Similarity" indicate areas of shared genetic material between different species.

Scoring Matrices

• Identity matrices assign a score of 1 for identical characters and 0 for non-identical ones.
• Similarity matrices consider the similarity between characters, allowing for scores that reflect the evolutionary relationship between them.
• PAM and BLOSUM matrices are examples of commonly used scoring matrices.

Pairwise Sequence Algorithms

• Optimal alignments can be constructed based on previous solutions for smaller subsequences.
• Dynamic programming is used to optimize the process, by calculating the best alignment that ends at a given pair of positions (i,j).

Key Elements of Sequence Alignment Algorithms

• Global pairwise alignment aims to align the entire length of two sequences, while local alignment identifies the best possible local alignment within the sequences.

Needleman-Wunsch Algorithm

• This algorithm produces an optimal global alignment between two sequences.
• It uses a 2D matrix to store partial alignment scores and iteratively calculates the best alignment.
• Scoring functions are used to assign values to matches, mismatches, and gaps.
• The actual alignment is constructed during the traceback phase of the algorithm, starting from the cell with the highest score and tracing back to the beginning.

Smith-Waterman Algorithm

• This algorithm focuses on local alignments, seeking the most significant region of similarity within two sequences.
• It modifies the Needleman-Wunsch algorithm to account for local alignment by setting negative scores to 0.
• The traceback process starts at the highest-scoring cell and ends when 0 is encountered.
• The algorithm can also be used to identify all local alignments that exceed a designated threshold.

Computational Complexity of Needleman-Wunsch Algorithm

• The algorithm has a computational complexity of O(n^2) for both speed and memory.
• Advanced versions can reduce memory usage to O(n), achieving linear space complexity.

Global vs. Local Alignment

• Global alignment seeks to align the entire lengths of two sequences.
• Local alignment identifies the best possible local region of similarity within two sequences.

Smith-Waterman Algorithm

• The initialization step sets the scores for the first row and column of the matrix to 0.
• The iteration step computes the score of each cell based on the maximum value obtained from three possible directions: alignment with a gap in the first sequence, alignment with a gap in the second sequence, or aligning the current characters.
• The termination and traceback phase involves identifying the cell with the highest overall score and tracing back to the starting point to reconstruct the optimal alignment.

Homework Assignment

• Students are required to complete a matrix for the Needleman-Wunsch algorithm and recover the corresponding alignment.
• They also need to create a "B" matrix in Python.

Recursion Example - Fibonacci Sequence

• The Fibonacci sequence is a numerical series where each number is the sum of the two preceding numbers.
• The sequence starts with 0 and 1.
• To find the Fibonacci sequence:
  • Add the last two digits to get the next digit.
  • Repeat the process to generate the remaining digits.

Dictionary Hash Tables

• Python's dict is a hash table data structure that efficiently stores key-value pairs.
• Dicts are created by using curly braces {} and are written as a series of key:value pairs, e.g., dict = {key1:value1, key2:value2, ...}.
• An empty dict is represented by an empty pair of curly braces {}.

Dict Access Operations

• Accessing or setting a value within a dict uses square brackets:
  • dict['foo'] looks up the value associated with the key 'foo'.
  • dict['foo'] = bar assigns the value bar to the key 'foo'.
• Keys in dicts can be strings, numbers, and tuples.
• Values can be of any data type.
• Attempting to access a non-existent key raises a KeyError.
• Use the in operator to check if a key exists: 'foo' in dict.
• The dict.get(key) method returns the value associated with the key, or None if the key is not present.
• The dict.get(key, not-found) method allows specifying a default value to return if the key is not found.

Example of Building a Dict

• A dict can be built incrementally by starting with an empty dict {} and adding key-value pairs using the assignment operator:
  • dict['a'] = 'alpha'
  • dict['g'] = 'gamma'
  • dict['o'] = 'omega'

Using Dictionaries as N-dimensional Arrays

• Dictionaries can function like N-dimensional arrays, storing values associated with multi-dimensional keys.
• The keys can be tuples or other composite data structures.
• Example: matrix[1, 2] = 5 assigns the value 5 to the key (1, 2).

Dict Iteration and Ordering

• By default, iterating over a dictionary (for key in dict: ...) iterates through its keys in an arbitrary order.
• Key-value pairs in a dictionary maintain the order in which they were originally inserted.
• dict.keys() returns a list of the keys.
• dict.values() returns a list of the values.
• dict.items() returns a list of (key, value) tuples, providing the most efficient way to access all the key-value data.
• These list methods' output can be sorted using the sorted() function.

Performance Advantages of Dicts

• Dictionaries provide a significant performance advantage for looking up and managing data based on keys.
• Utilize dictionaries when possible to organize and access data efficiently.

Incremental Development (Rapid Prototyping) in Python

• Don't write entire Python programs at once.
• Identify smaller milestones and write code to reach them one at a time.
• Use print statements to inspect data structures at each milestone.
• Use sys.exit(0) to halt program execution after a milestone to focus on specific areas.
• This gradual approach allows for quick testing and iteration, making it easier to build and refine complex programs.

Introduction to Lab 4 - Creating a Dictionary of Codons and Amino Acids

• The lab focuses on building a dictionary representing the mapping between codons (three-nucleotide sequences) and their corresponding amino acids.
• You will complete a dictionary (aminos) containing this information.
• The lab provides code to translate DNA sequences into amino acid sequences using the aminos dictionary.

Lab 4 Key Concepts:

• The code includes a function pick_a_codonthat randomly selects a codon for a given amino acid (if there are multiple codons that map to the same amino acid).
• The lab emphasizes understanding the structure and manipulation of dictionaries in Python for tasks like translation (DNA -> Amino Acid).
• It also introduces concepts like open reading frames (sequences that are translated into proteins until a stop codon is encountered).