STBC2023

1
STBC2023 Introduction to
BioinformaticsAnalyses Predictive Methods
Using Nucleotide Sequences

• M. Firdaus Raih
• Room 1166, Bangunan Sains Biologi
• Office Hours Wednesdays
• Phone [phone] Email firdaus_at_mfrlab.org

Ver. 23-01-09-1
2
Guide

• This is a electronic self study and self
  assessment module which is based on the lectures
  which cover Topic 4 Analysis at Nucleotide
  Level of the STBC2023 Introduction to
  Bioinformatics course.
• To navigate this module, use the buttons provided
  mostly on the bottom right hand corner of the
  page or in some slides, the bottom left hand
  corner. The Home icon button will automatically
  set the slide back to the key questions which we
  are trying to answer with this course material.
  Several pages have hyperlinks which navigate
  immediately to either specific slides OR navigate
  away from this module via the default web
  browser. To return, simply click back this file.
  Not clicking on the buttons properly will result
  in normal powerpoint slideshow mode progression
  of the slides as opposed to navigating to the
  directed pages.
• Practicals and self assessment questions to gauge
  your comprehension of a given practical session
  are also provided throughout. Please attempt the
  practicals and the questions on your own before
  resorting to the solutions or answers provided.

3
Pre-session Questions

• What are nucleic acids?
• What types of nucleic acids are there?
• What functions do nucleic acids have?
• What sort of information do nucleotide sequences
  carry?
• What can be done with DNA sequences?
• What can be done with RNA sequences?
• Is molecular structure important for RNA
  sequences?
• What is a sequence alignment?
• What is the relationship of an alignment with
  regard to biological function?
• Is extracting the encoded information for protein
  synthesis the only sequence analysis which can be
  done?

4
Learning objectives

• Know the basic chemistry and able to understand
  the diverse functions of nucleic acids.
• Able to generally list potential analyses for
  nucleic acid sequence data and the applications
  for those analyses based on an understanding of
  the functions of nucleic acids.
• Able to formulate a strategy and present
  processes involved in the analysis of nucleic
  acid sequence data.
• Able to comprehend the basic concepts involved in
  sequence alignments in general and aligning
  nucleic acids specifically as well as the
  relationship between an alignment to a sequences
  biological function.

5
Nucleic Acids Chemistry and Molecular Structure

• What are nucleic acids?

6
Nucleic Acids Chemistry and Molecular Structure

• What are nucleic acids?
• Nucleic acids polymer of nucleotides ? 2 types
• DNA deoxyribonucleic acids
• RNA ribonucleic acids

7
Nucleic Acids Chemistry and Molecular Structure

• What are nucleic acids?
• Nucleic acids polymer of nucleotides ? 2 types
• DNA deoxyribonucleic acids
• RNA ribonucleic acids
• What is a nucleotide?

8
Nucleic Acids Chemistry and Molecular Structure

• What are nucleic acids?
• Nucleic acids polymer of nucleotides ? 2 types
• DNA deoxyribonucleic acids
• RNA ribonucleic acids
• What is a nucleotide?
• Nucleotide nucleoside 1 phosphate group
• Nucleoside nitrogenous base sugar (ribose)

9
Nucleic Acids Chemistry and Molecular Structure
What is the basic difference between RNA and DNA
(in terms of chemistry)?
RNA
DNA
Click here for animation
10
Nucleic Acids Chemistry and Molecular Structure
How can the nucleotide polymer be represented?
11
Nucleic Acids Chemistry and Molecular Structure
How can the nucleotide polymer be represented?
Hydrogen bonded base interactions and base
stacking interactions result in stable structures
of DNA / RNA.
Seq 1. ACTG Seq 2. TGAC
What can be done with such sequence data? How is
the analysis related to biological function?
12
Nucleic Acids Biological Functions

• What is/are the function(s) of DNA?
• What is/are the function(s) of RNA?

13
Nucleic Acids Biological Functions

• What is the function of DNA?
• Storage of genetic information
• Proteins such as transcription factors also
  interact directly with DNA as part of regulatory
  pathways
• Total genetic content of an organism genome
• Genes are part of genomes
• So what is a gene?

14
Nucleic Acids Biological Functions

• What is the function of DNA?
• Storage of hereditary information in genes.
• What is a gene?

While sequencing of the human genome surprised us
with how many protein-coding genes there are, it
did not fundamentally change our perspective on
what a gene is. In contrast, the complex patterns
of dispersed regulation and pervasive
transcription uncovered by the ENCODE project,
together with non-genic conservation and the
abundance of noncoding RNA genes, have challenged
the notion of the gene. To illustrate this, we
review the evolution of operational definitions
of a gene over the past century--from the
abstract elements of heredity of Mendel and
Morgan to the present-day ORFs enumerated in the
sequence databanks. We then summarize the current
ENCODE findings and provide a computational
metaphor for the complexity. Finally, we propose
a tentative update to the definition of a gene A
gene is a union of genomic sequences encoding a
coherent set of potentially overlapping
functional products. Our definition side-steps
the complexities of regulation and transcription
by removing the former altogether from the
definition and arguing that final, functional
gene products (rather than intermediate
transcripts) should be used to group together
entities associated with a single gene. It also
manifests how integral the concept of biological
function is in defining genes.
15
Nucleic Acids Biological Functions

• What are the functions of RNA?

16
Nucleic Acids Biological Functions

• What are the functions of RNA?
• Information storage and transfer
• Genomes of RNA viruses
• mRNA
• Protein synthesis
• tRNA
• Peptidyl transferase
• Catalysis
• ribozymes
• Regulatory
• Small ncRNAs / microRNAs
• Riboswitches
• Also see The RNA World hypothesis first coined
  by Walter Gilbert 1986, Nature

17
DNA (Genes) From Sequence to Function

• How does a gene sequence correlate to biological
  function?

18
DNA (Genes) From Sequence to Function

• How does a gene sequence correlate to biological
  function?
• Lets first look at
• Information about the amino acid sequence is
  contained within the nucleic acids sequence.
• Is that the only analysis that can be done for
  DNA sequences?
• What other analyses, if any, can be done for DNA
  sequences?

19
Potential Analyses for DNA Sequences

• What can be done with DNA sequences?
• Genome projects DNA sequencing data need to be
  assembled into complete genomes.
• Genes need to be identified / predicted.
• Comparisons of specific nucleotide level
  variations.
• Identification and analysis of specific
  nucleotide sequence level motifs and patterns.
• Identification and analysis of polymorphisms.

20
Potential Analyses for DNA Sequences

• What can be done with DNA sequences?
• Genome projects DNA sequencing data need to be
  assembled into complete genomes.
• Genome sequencing generate fragments of sequences
  .
• These fragments need to be assembled into genes,
  chromosomes and finally the complete genome.
• Assembly is done by analyzing for contiguous
  sequences (contigs).
• Contigs are basically found by aligning the short
  DNA sequences to one another and finding where
  there are overlaps.
• More on this topic will be covered in the
  Genomics course in Year 3.
• After the genome is assembled, the genes need to
  be identified.

21
Potential Analyses for DNA Sequences

• What can be done with DNA sequences?
• From sequence data, genes need to be predicted.
• Several methods to gene prediction
• Searching by signal analysis of sequence
  signals which specify a gene.
• Searching by content analysis of regions
  showing compositional bias that has been
  correlated to coding regions.
• Homology based prediction comparison against
  known gene sequence. involve sequence
  alignments
• Comparative gene prediction comparing sequences
  of interest against anonymous genomic sequences.
  involve sequence alignments
• The prediction of eukaryotic genes from genomic
  DNA data is appreciably more difficult than that
  of prokaryotic. Why?

22
Potential Analyses for DNA Sequences

• What can be done with DNA sequences?
• From sequence data, genes need to be predicted.
• Several methods to gene prediction
• Searching by signal analysis of sequence
  signals which specify a gene.
• Searching by content analysis of regions
  showing compositional bias that has been
  correlated to coding regions.
• Homology based prediction comparison against
  known gene sequence. involve sequence
  alignments
• Comparative gene prediction comparing sequences
  of interest against anonymous genomic sequences.
  involve sequence alignments
• For this session, we will focus on methods which
  involve sequence alignments.

23
Potential Analyses for DNA Sequences

• What can be done with DNA sequences?
• Comparisons of specific nucleotide level
  variations.
• Enable differentiation at individual level or
  close relationships ie. Between strains of the
  same species.
• Phylogenetic analysis (discussed by Dr.
  Khairina).

24
Potential Analyses for DNA Sequences

• What can be done with DNA sequences?
• Identification and analysis of specific
  nucleotide sequence level motifs, patterns.
• This will be discussed further in the following
  lecture.
• Examples
• PCR Primer design
• Searching / mapping restriction sites
• Go to the corresponding BLAST exercise NOW
• or proceed to the next slide.

25
Potential Analyses for DNA Sequences

• What can be done with DNA sequences?
• Identification and analysis of polymorphisms.
• This will be discussed further in the following
  lecture.
• Examples
• SNPs single nucleotide polymorphisms (more on
  SNPs)
• Go to the corresponding BLAST exercise NOW
• or proceed to the next slide.

26
Sequence Alignments

• What is a sequence alignment?
• A way of arranging or aligning the
  similarities between sequences.
• Examples
• Gaps (-) are inserted to optimize alignments.
• They represent indel mutations.
• Easy to align short sequences manually. But what
  about longer sequences? How can those be aligned?
  In order to understand this further, lets look
  at a method which we can visualize and track the
  alignment. This method is called a dot plot.

27
Sequence Alignments

• What is a dot plot?
• A plot where two sequences are written along the
  top row and leftmost column of a two-dimensional
  matrix and a dot is placed at any point where the
  characters in the appropriate columns match.
• Parts of the two sequences where the match is
  continuous can be traced as a diagonal line ?
  region where the sequences are aligned.
• A sequence can be plotted against itself and
  regions that share significant similarities will
  appear as lines off the main diagonal can occur
  when a protein consists of multiple similar
  structural domains.
• A dot plot is not able to detect divergence or
  substitutions/mutations which we know can occur.

28
Sequence Alignments

• Dot plot for two DNA sequences
• Complete the dot plot for the two DNA sequences
  provided below. An example can be seen below.
• Seq1 CGATCGCGTAATCGGTGATCGGC
• Seq2 CGGTATCGGTGATCGATCGCA
• Questions
• Which stretch of these sequences can best be
  aligned to each other? (Answer)
• Can this alignment be extended? (Answer)
• Can you identify a repetitive sequence of 4 bases
  which keep occurring in both sequences? (Answer)

29
Sequence Alignments

• Dot plot for two DNA sequences
• Questions
• 1. Which stretch of these sequences can best be
  aligned to each other?
• Answer Longest continuous diagonal line from
  your dot plot. ATCGGTGATCG
• 2. Can this alignment be extended?
• Answer Yes, it can be extended as shown below. 2
  nucleotides are not aligned and may possibly be
  substitutions.
• 3. Can you identify a repetitive sequence of 4
  bases which keep occurring in both sequences?
• Answer ATCG, this can be deduced from the
  repeating short diagonal lines.
• 4. What can you attribute all the other plotted
  dots to?
• Answer They are the result of random sequence
  similarities.

30
Computational Sequence Alignments

• Weve looked at manual alignments for short
  sequences and the dot plot However, manual
  alignments cannot be done for lengthy and highly
  variable sequences. Therefore for long variable
  sequences, computer aided alignments need to be
  done.
• How can computer aided alignments be done?
• To enable computer aided alignment, algorithms
  called dynamic programming algorithms are used.
• Two common dynamic programming algorithms
  approach alignment differently, via
• 1. Local alignments Smith-Waterman algorithm
• 2. Global alignments Needleman-Wunsch algorithm

31
Computational Sequence Alignments

• What is the difference between global and local
  alignments?
• Local alignments Smith-Waterman algorithm
• Global alignments Needleman-Wunsch algorithm
• The Smith-Waterman algorithm is currently the
  most used because real biological sequences are
  usually similar in localized portions and not
  over entire lengths.
• Examples
• genes from different organisms with similar
  exons, different intron structures
• Proteins share only certain domains
• Alignments can have gaps which represent
  mutations. The ability to add gaps is required as
  sequence diverge.
• So how do we know that an alignment is
  meaningful?

32
Computational Sequence Alignments

• How do we know that an alignment is meaningful?
• Insertions and deletions are slow evolutionary
  processes, therefore addition of gaps MUST be
  controlled to avoid large proportions of matches
  by inserting large numbers of gaps.
• Gap penalties are given to control addition of
  gaps. The penalty system can be constant or
  proportional.
• Scores are given for matches, while penalties are
  given for addition of gaps.
• The alignment algorithm then carries out
  alignments in order to get the best score.
• Like the dot plot, a simple system as above does
  not seem to fully consider divergence (ie. point
  mutations) only deletions and insertions seem
  to be considered.
• How can we get around this problem?

33
Computational Sequence Alignments

• How do we know that an alignment is meaningful?
  (cont.)
• Point mutations can result in change as opposed
  to deletion or insertion.
• A matrix called a substitution matrix can be used
  to model the possible changes and provide
  quantitative values to changes arising from point
  mutations.
• The values for substitution can
• take into consideration similarity
• such as physico-chemical
• properties for amino acids or
• transition mutations for nucleic
• acids.
• But there is still probability
• that a search result is random
• especially for large databases.
• How can we be certain the
• alignment achieved is the
• expected result?

Amino acid substitution matrix example
Nucleic acid substitution matrix example
34
Computational Sequence Alignments

• How do we know that an alignment is meaningful?
  (cont.)
• How can we be certain the alignment achieved is
  the expected result?
• The alignments produced are statistically
  evaluated.
• As an example, for the BLAST program, a value
  called the Expectation (E) value is given.
• The number of different alignments with scores
  equivalent to or better than S that are expected
  to occur in a database search by chance.
• The lower the E value, the more significant the
  score.

35
Sequence Alignments

• What is the rationale in doing an alignment?
• Proteins perform most cellular functions.
• The structure of a protein is an important
  determinant of its function.
• If proteins share a similar structure, then it
  may also share a similar function.
• We know that sequences with 30 similarity, share
  a similar fold (Chothia Lesk 1986).

36
Sequence Alignments

• What is the rationale of doing an alignment?
• If proteins share a similar function, then it may
  also share a similar structure.

37
Sequence Alignments

• What is the rationale in doing an alignment?
• If proteins share a similar structure, then it
  may also share a similar sequence.
• But our interest here are NUCLEIC ACID sequences
• So what is the relevance?

38
Sequence Alignments

• What is the rationale in doing an alignment?
• If proteins share a similar structure, then it
  may also share a similar sequence.
• But our interest here are NUCLEIC ACID sequences
• So what is the relevance?

39
Sequence Database Searching

• What is a sequence database?
• What are we searching for, and how do we search
  for something in sequence databases?

40
Sequence Database Searching

• What is a sequence database?
• A collection of biological macromolecular
  sequences.
• Can be sequences organized into organisms,
  protein families, sources etc.
• Has been covered in Topic 2. Example NCBI
  GenBank.
• What are we searching for, and how do we search
  for something in sequence databases?

41
Sequence Database Searching

• What is a sequence database?
• A collection of biological macromolecular
  sequences
• Can be sequences organized into organisms,
  protein families, sources etc.
• Has been covered in Topic 2. Example NCBI
  GenBank
• What are we searching for, and how do we search
  for something in sequence databases?
• We are searching for sequence similarity.
• We can search for sequence similarity by
  comparing an input (query) sequence against
  sequences in the database.
• This comparison is done by aligning the query
  sequences to the database sequences ? one tool we
  can use is BLAST.
• How is this alignment relevant biologically?

42
Sequence Database Searching

• What is BLAST?
• Basic Local Alignment Search Tool
• Implements heuristics to approximate the
  Smith-Waterman algorithm and search for high
  scoring alignments.
• The alignment scores are then statistically
  evaluated one example is the E value discussed
  previously.
• BLAST is actually a family of programs.

43
Sequence Database Searching

• What is BLAST?
• Basic Local Alignment Search Tool
• Implements heuristics to approximate the
  Smith-Waterman algorithm and search for high
  scoring alignments.
• The alignment scores are then statistically
  evaluated one example is the E value discussed
  previously.
• BLAST is actually a family of programs.

44
BLAST

• How do we use BLAST?

45
BLAST

• How do we use BLAST?

(1) Select the BLAST program (2) Input the
sequence (query) (3) Choose the database to
search (4) Choose optional parameters Then
click BLAST
46
BLAST

• Is that it?

47
BLAST

• Is that it?... YES and NO. Lets look at some
  considerations and strategies for BLAST
  searching.

48
BLAST

• Some considerations and strategies
• Input sequence and search database what is it
  that youre really interested in? Finding
  similarity alone or identifying homologs? Finding
  homologs only or perhaps trying to find out if
  genes with similar sequences encode for proteins
  with available structures? The answer to these
  types of questions influence the type of search
  program you should use and the database to search
  in.

Protein vs. Nucleotide?
49
BLAST

• Some considerations and strategies
• Are you interested in something quite specific?

50
BLAST

• Some considerations and strategies
• Did you forget to turn something on/off?
• Sequence filters Low-complexity regions have
  fewer sequence characters in them because of
  repeats of the same sequence character or
  pattern. These sequences produce artificially
  high-scoring alignments that do not accurately
  convey sequence relationships in sequence
  similarity searches. Regions of low complexity or
  repetitive sequences may be readily visualized in
  a dot matrix analysis of a sequence against
  itself. Low-complexity regions with a repeat
  occurrence of the same residue can appear on the
  matrix as horizontal and vertical rows of dots
  representing repeated matches of one residue
  position in one copy of the sequence against a
  series of the same residue in the second copy.
  Repeats of a sequence pattern appear in the same
  matrix as short diagonals of identity that are
  offset from the main diagonal. Such sequences
  should be excluded from sequence similarity
  searches.

51
BLAST

• Some considerations and strategies
• Did you forget to turn something on/off?
• Options and parameter settings

52
Output of BLAST Searches

• What are the components of a BLAST search output
• Example blastn vs blastx (GenBank AF390557)

blastn
This section overview of the output alignments
blastx
53
Output of BLAST Searches

• What are the components of a BLAST search output
• Example blastn vs blastx (GenBank AF390557)

blastx
blastn
This section list of hits (alignments) Read
more about interpreting the output.
54
Output of BLAST Searches

• What are the components of a BLAST search output
• Example blastn vs blastx (GenBank AF390557)

blastx
blastn
This section the alignments
55
Output of BLAST Searches

• To be a significant match, a database sequence
  that is listed in the program output should have
  a small E (expect value) and a reasonable
  alignment with the query sequence (or
  translations of protein-encoding DNA sequences
  should have these same features).
• The E of the alignment score between the
  sequences gives the statistical chance that an
  unrelated sequence in the database or a random
  sequence could have achieved such a score with
  the query sequence, given as many sequences as
  there are in the database. The smaller the E, the
  more significant the alignment. A cutoff value in
  the range of 0.01-0.05 may be used (Pearson
  1996). In genome comparisons, a more stringent
  cutoff score (10-100-10-20) may be used to find
  sequences that align very well with the query
  sequence. However, the alignment should also be
  examined for absence of repeats of the same
  residue or residue pattern because these patterns
  tend to give false high alignment scores.
• Filtering of low-complexity regions from the
  query sequence in a database search helps to
  reduce the number of false positives. The
  alignment should also be examined for reasonable
  amino acid substitutions and for the appearance
  of a believable alignment.
• To gain further confidence that the alignment
  between the query and database sequences is
  significant, either the query sequence or the
  matched database sequence may be shuffled many
  times, and each random sequence may be realigned
  with the other unshuffled sequence to obtain a
  score distribution for a set of unrelated
  sequences. This distribution may then be used to
  evaluate the significance of the true alignment
  score.

.
56
BLAST

• Carrying out a BLAST search
• Select and copy the sequence from the GenBank
  database here.
• Go to the BLAST page and carry out database
  searches using the above sequence.
• First carry out a search against a nucleotide
  database.
• Which BLAST programs can you use? Name two
  possibilities. (Answer)
• Next carry out a search against a protein
  database
• Which BLAST program should you use? (Answer)
• (i) Can you further narrow down the search? (ii)
  Also take for example if you were to search for
  genes which code for proteins which have
  representative 3D structures how would you
  conduct such a search? (Answer)

57
BLAST

• Answers to questions on carrying out a BLAST
  search
• First carry out a search against a nucleotide
  database.
• Which BLAST programs can you use? Name two
  possibilities.
• Answer blastn and tblastx. tblastn is not a
  correct answer because it uses a protein query
  although the database searched is a nucleotide
  database the input sequence AF390557 is a DNA
  sequence.
• Next carry out a search against a protein
  database
• Which BLAST program should you use?
• Answer blastx
• (i) Can you further narrow down the search? (i)
  Also take for example if you were to search for
  genes which code for proteins which have
  representative 3D structures how would you
  conduct such a search?
• Answer (ii) Yes, searches returning a very large
  number of hits can still be narrowed down. A
  carefully annotated protein sequence database
  (e.g., PIR, SwissProt) will provide a more
  manageable output list of matched sequences, and
  these proteins have probably been observed in the
  laboratory i.e., the genes do produce a protein
  product in cells. However, investigators may also
  wish to expand the search to include predicted
  genes from gene annotations of genomic sequences
  that are frequently entered into the DNA sequence
  translation databases (e.g., DNA sequences in the
  GenBank DNA sequence databases automatically
  translated into protein sequences and placed in
  the GenPept protein sequence database). To
  compare a protein or predicted protein sequence
  to EST sequences, the ESTs should be translated
  into all six possible reading frames. (ii) Such a
  search can be carried out by choosing PDB as the
  database option. This will limit the blastx
  search to only protein sequences which have known
  3D structures in the PDB.

58
BLAST

• Carrying out a BLAST search
• Retrieve the sequence provided and use it for
  your BLAST search.
• See the GenBank page here for the sequence.
  Change the format of the view to FASTA by
  selecting FASTA from the dropdown menu marked
  Display (see here). Use this sequence for a
  BLAST search.
• Questions
• - Identify the sequence which is used. What is
  this DNA usually used for? (Answer)
• - Search for suitable primers to use for PCR.
  Which program can you use? (Answer)
• - Identify restriction sites which can be found
  on this DNA. How many fragments will a digestion
  with the restriction enzyme BsaI generate? In
  order to answer this question, you will need to
  draw on any general web skills you already have
  to find the appropriate resources. BLAST is not
  the tool to use in such a case. (Answer)

59
BLAST

• Carrying out a BLAST search
• Questions
• - Identify the sequence which is used. What is
  this DNA usually used for?
• Answer pBR322 plasmid, It is used a cloning
  vector for protein (IG-lambda) expression.
• - Search for suitable primers to use for PCR.
  Which program can you use? What is the largest
  product size from a possible primer pair found
  using a default search?
• Answer The Primer-BLAST program can be used.
  The largest possible product is 986bp.
• - Identify restriction sites which can be found
  on this DNA. How many fragments will a digestion
  with the restriction enzyme BsaI generate?
• Answer One such tool which can be used is
  NEBcutter. Cutting the pBR322 sequence with BsaI
  will generate 3 fragments of DNA due to cleavage
  at 2 sites in the sequence.

60
SNPs

• SNPs (pronounced snips) is a DNA sequence
  variation which occurs when a single nucleotide
  A, T, C, or G in the genome (or other shared
  sequence) differs between members of a species
  (or between paired chromosomes in an individual)
  and they comprise the largest known class of
  human genetic variation.
• SNPs may occur
• within coding sequences of genes,
• non-coding regions of genes, or
• in the intergenic regions between genes.
• SNPs within a coding sequence will not
  necessarily change the amino acid sequence of the
  protein that is produced, due to degeneracy of
  the genetic code (refer to the codon table
  discussed earlier) such changes result in silent
  mutations (synonymous).
• Non-synonymous changes can result in
• Mis-sense change ? different amino acid coded
• Nonsense change ? premature STOP codon
• Why are SNPs important? If the changes result in
  non-functional gene products or no gene products,
  a diseased state may be a possible the end
  result.
• How can we find SNPS? Methods of discovering SNPs
  in sequence data the easiest and most used
  method is to align two sequences from the DNA of
  two individuals and look for high quality
  sequence differences.

.
61
BLAST

• Carrying out a BLAST search
• Select and copy the sequence from this link.
• Go to the BLAST page and carry out a search for
  SNPs on the above sequence.
• Observe the output. How is it different from
  previous BLAST searches you have carried out.
  Correlate the output to what you know about SNPs.
  

62
Ribonucleic Acids

• RNA molecules play crucial roles in molecular
  biology.
• Known functions include
• Information storage
• Catalysis
• Regulatory roles
• Protein synthesis
• Diversity of functions associated to RNA World
  hypothesis
• Potential applications
• Molecular scaffolding (nanotechnology)
• Drug targets (riboswitches/ribosomes)
• RNA interference (RNAi)

The Economist, June 16th-22nd 2007
63
RNA From Sequence to Function

• What is a crucial determinant of functionality
  for functional RNAs?

64
RNA From Sequence to Function

• What is a crucial determinant of functionality
  for functional RNAs?
• For functional RNAs, like for proteins, the 3D
  structure is crucial for biological function.

65
RNA Structure

• What are the major factors involved in
  stabilizing the structure of RNA?
• Base stacking and hydrogen bonding contribute to
  the stabilization of nucleic acid structure/ RNA
  structure.
• RNA bases can form hydrogen bonds with each other
  resulting in interactions between
• complementary pairings in the canonical Watson
  Crick interactions
• non-canonical interactions
• Hydrogen bonded base interactions are therefore
  are crucial elements of a nucleic acids 3D
  structure.

66
RNA Base Interactions

32 pairs
eg. Purine-pyrimidine base pairs (10)
after I. Tinoco, Jr. In Appendix 1 of The RNA
World (R. F. Gesteland, J. F. Atkins, Eds.),
Cold Spring Harbor Laboratory Press, 1993, pp.
603-607.
67
RNA Structure

• Base stacking and hydrogen bonding contribute to
  the stabilization of nucleic acid structure/ RNA
  structure.
• RNA bases can form hydrogen bonds with each other
  resulting in interactions between
• complementary pairings in the canonical Watson
  Crick interactions
• non-canonical interactions
• Hydrogen bonded base interactions are therefore
  are crucial elements of a nucleic acids 3D
  structure
• 3 levels of RNA structure
• Primary sequence, secondary structure, tertiary
  structure.

from the Arabic word Qanun which in context
here is better suited as the word rule as
opposed to the literal meaning of law.
68
RNA Structure

• How do we get from sequence to structure?
• How can we predict the structure of RNA?

69
RNA Structure

• How do we get from sequence to structure?
• Complex (non helical) RNA structures are not easy
  to predict. Reliable structural information are
  sourced from X-ray crystal structures.
• Commonly, only the secondary structure level
  interactions are predicted to give some insights
  into what the functional structure may look like.
• However such methods lack the detail which an
  actual structure model is able to give, such as
  the exact orientation of bases and specific
  atomic interactions which are occurring.
• Such interaction data is important because we
  know that RNA bases can be involved in
  non-canonical interactions which are different
  from the canonical Watson-Crick interactions.
• How can we predict the secondary structure of
  RNA?
• Several programs which calculate the
  thermodynamics of folding (energies of the base
  interactions) can be used.
• One such program is mfold by Michael Zuker.
• Assessment of reliability can be done using
  multiple alignments and comparisons to other
  predictions and known structures.

70
RNA Secondary Structure Prediction the mfold
program

• Predicting the secondary structure of non-coding
  RNA
• Copy the sequence here as input for the mfold
  program. All other parameters can be left at
  default settings.
• Questions
• How many paired bases are you able to observe in
  the predicted structure? (Answer)
• How many bases are unpaired? (Answer)
• Name the two types of structures where these
  unpaired bases can be found. What type of
  secondary structure do you think can be observed
  for regions with canonical Watson-Crick base
  pairing? (Answer)
• Are you able to observe any base pairings which
  are non-canonical (non Watson-Crick)? If yes, how
  many? (Answer)
• Having answered the previous two questions, are
  you really able to differentiate a canonical vs a
  non-canonical pairing from the secondary
  structure diagram alone? (Answer)

71
RNA Secondary Structure Prediction the mfold
program

• Predicting the secondary structure of non-coding
  RNA
• How many paired bases are you able to observe in
  the predicted structure?
• Answer 29 pairs, 58 paired bases.
• How many bases are unpaired?
• Answer 27
• Name the two types of structures where these
  unpaired bases can be found. What type of
  secondary structure do you think can be observed
  for regions with canonical Watson-Crick base
  pairing?
• Answer Unpaired bases are found in bulges and
  loops. Regions with canonical pairings as in
  Watson-Crick are most likely helical.
• Are you able to observe any base pairings which
  are non-canonical (non Watson-Crick)? If yes, how
  many?
• Answer 4
• Having answered the previous two questions, are
  you really able to differentiate a canonical vs a
  non-canonical pairing from the secondary
  structure diagram alone?
• Answer No, not really. Although a GU base pair
  is obviously non-canonical, GC and AU base pairs
  which may possibly be non-canonical cannot be
  determined from the secondary structure alone.

72
Analyses for RNA sequence data

• Is predicting the secondary structure the only
  analyses we can do for RNA sequence data?

73
Analyses for RNA sequence data

• Is predicting the secondary structure the only
  analyses we can do for RNA sequence data?
• NO.
• Genomic data can be analysed for the presence of
  the numerous types of known non-coding or
  functional RNA as well as possibly novel or yet
  to be discovered functional RNA sequences.
• This appreciably more difficult than the problem
  of predicting genes. Why?
• Currently there are no widely used or general use
  methods.
• Such investigations are still highly exploratory
  and currently remain in the domain of experts in
  the field.

74
Post-session Questions

• What are nucleic acids?
• What types of nucleic acids are there?
• What functions do nucleic acids have?
• What sort of information do nucleotide sequences
  carry?
• What can be done with DNA sequences?
• What can be done with RNA sequences?
• Is molecular structure important for RNA
  sequences?
• What is a sequence alignment?
• What is the relationship of an alignment with
  regard to biological function?
• Is extracting the encoded information for protein
  synthesis the only sequence analysis which can be
  done?

75
Self Study and Self Assessment

• The self study module for this series of lectures
  on analyses of nucleotide sequences are available
  for download from SPIN. Format of the file (this
  file) is powerpoint show (.pps).
• The self assessment quiz is accessible from
  within the SPIN interface.
• Both these materials are for self assessment and
  self study use and DOES NOT contribute to your
  final grades for this course.
• Also explore the references and texts listed in
  the course information file and reading list.
• Explore resources made available via the
  self-study material.

76
Further Reading

• Recommended Textbook (Lesk, 2nd Ed.)
• Basics Chapter 1
• Pages 1-59
• Sequence alignments Chapter 5, Chapter 1
• Pages 242-270
• Pages 21-59
• Other Textbooks
• Baxevanis Oullette, 3rd edition
• Chapters 5-7
• Pevsner