An Introduction to Bioinformatics

1
An Introduction to Bioinformatics

• Laboratory of Computational Molecular Biology
• College of Life Sciences
• Beijing Normal University

2
Image from http//microscopy.fsu.edu
3
The Timeline of Landmark Accomplishments in
Genetics and Genomics

• 1866 - Gregor Mendel's discovery of the laws of
  heredity
• 1944 - Recognition of DNA as the hereditary
  material
• 1953 - Determination of DNA structure
• 1963 - Elucidation of the genetic code
• 1972/3 - Development of recombinant DNA
  technologies
• Establishment of increasingly automatable methods
  for DNA sequencing (1975,1977,1986, 2001)
• 1990 - The stage for the Human Genome Project
  (HGP)
• 2001 - The IHGSC and Celera Genomics each
  reported draft sequences providing a first
  overall view of the human genome
• 2005 - Human haplotype map (HapMap)

4
Genome Sequencing Technology

• Traditional clone-by-clone approach
• High-throughput whole genome shotgun (WGS)
  sequencing assembly
• Advance in assembly algorithms
• Challenges repetitive sequences, diploid genome,
  etc. are difficult for correct assembly
  therefore more robust automated methods are
  needed.

Next-generation sequencing outpaces expectations
Nature Biotechnology
2007. Vol 25149. Several orders of magnitude
more efficient than Sanger capillary-array
electrophoresis (CAE) machines. 1,000 genome
program 600 megabases per day -gt 100
billion bases per day Targeting resequencing
efforts aimed at finding genetic variations
and rare mutations that contribute to complex
diseases.
5
Why is There Bioinformatics?

• High-throughput molecular biological technologies
• Automated sequencers
• Genomes, ESTs, SNPs, etc.
• DNA array for large-scale gene expression
• Proteomics platform
• MS/MS LC/MS/MS (liquid chromatography),
  Protein-protein interaction experiments,
  genome-wide localization technique
• Massive datasets produced

• GenBank (As of August 2006)
• 65,369,091,950 bases in 61,132,599 sequence
  records in the traditional GenBank divisions
• 80,369,977,826 bases in 17,960,667 sequence
  records in the WGS division.

6
Is Biology an Informational Science?

• The HGP changed how we view practice biology.
• Biology is an informational science.
• Digital genome
• Environmental signals
• Biology has become a cross-disciplinary science.

From U.S. Department of Energy Human Genome
Program. http//www.ornl.gov/hgmis
7
Bioinformatics as an intersecting discipline
Developing the high throughput technologies and
computational/mathematical tools required for
this new biology.
8
Why? Where? What? How?

• Why Ideas for what to produce these huge
  datasets? Biological background needed.
• Where Raw data need to store, IT platforms
  required.
• What Patterns in datasets that can be analyzed
  using computers. Various data models and their
  respective algorithms are needed.
• How Different resources need to be integrated.

9
What is Bioinformatics?

• The field of biology specializing in developing
  hardware and software to store and analyze the
  huge amounts of data being generated by life
  scientists. (NIH)
• More than 20 different definitions can be found
  from Google!

Bioinformatics applications Data analysis
Data integration Various molecular biology
databases
10
Key Challenge of Bioinformatics

• The world of biology is very different from what
  it was even ten years ago.
• To bridge the considerable gap between technical
  data production and its use by scientists for
  biological discovery.

11
Main resources for Bioinformatics

• Databases
• GenBank, EMBL, DDBJ
• UniProt/SwissProt, InterPro, Pfam, SCOP, PDB
• Gene Ontology, KEGG
• Algorithms Applications
• BLAST, FASTA, BLAT
• ClustalW, HMMer
• Phylip

12
(No Transcript)
13
1138 Prokaryotic Genome Sequencing Projects
Selected Complete - 461, Draft assembly - 307,
In Progress 370 257 Eukaryotic Genome
Sequencing Projects Selected Complete - 24,
Draft assembly - 95, In Progress 138 (NCBI
2007-2-27, http//www.ncbi.nlm.nih.gov/genomes/sta
tic/gpstat.html)
14
NCBIs Tools for Data Mining

• Nucleotide Sequence Analysis
• Model Maker - allows you to view the evidence
  (mRNAs, ESTs, and gene predictions) that was
  aligned to assembled genomic sequence to build a
  gene model and to edit the model by selecting or
  removing putative exons.
• Protein Sequence Analysis and Proteomics
• CD Search - search the Conserved Domain Database
  with Reverse Position Specific BLAST.
• Structures
• Genome Analysis
• Entrez Genomes - whole genomes of over 1000
  organisms. It provides graphical overviews of
  complete genomes/chromosomes and the ability to
  explore regions of interest in progressively
  greater detail.
• Gene Expression
• The Cancer Genome Anatomy Project - aims to
  decipher the molecular anatomy of cancer cells.
  CGAP develops profiles of cancer cells by
  comparing gene expression in normal,
  precancerous, and malignant cells from a wide
  variety of tissues.

15
What are the Hot Topics in Bioinformatics?

• Comparative genomics
• Genetic variation analysis
• Microarray-based gene
• expression analysis
• Systems biology

The idea that we can study the interactions of
all elements in a biological system and from
these come to understand its emergent properties.
Gene ? Genome all of the genes in an
organism Transcipt ? Transcriptome expressed
in a particular biological process Protein ?
Proteome the world of expressed
proteins Interaction ? Interactome the
interactions between molecules
Genomics, Transcriptomics, Proteomics,
Metabolomics/Interactomics
16
Browsers Light applications
Clients
Intranet and/or Internet
HTML PERL/C/C/Java MySQL Bioinformatics
applications HPC with MPI
WWW servers Database servers Intensive computing
servers
Servers
17
The future of genomics research Nature 422
835-847.2003
.
18

• Nothing in biology makes sense except in the
  light of evolution.
• ------ Theodosius Dobzhansky

The evolutionary process finds a way to create
exceptions to every model we propose. --- Austin
L. Hughes
19
Biological Data Objects
Genomes Protein sequences

• Laboratory of Computational Molecular Biology
• College of Life Sciences
• Beijing Normal University

20
Sequence assembly by the shotgun approach
Building up the master sequence directly from the
short sequences obtained from individual
sequencing experiments, simply by examining the
sequences for overlaps. This is called the
shotgun approach. It does not require any prior
knowledge of the genome and so can be carried out
in the absence of a genetic or physical map.

21
So what are you going to do with this?
How do you find where the genes are?
CGGTTGAAAGCGGTAGCGTCCATGCGTATTACTCTTGAGCGGTCGAACCT
TCTGAAATCGCTGAACCACGTCCACCGGGT CGTCGAGCGTCGCAACACG
ATCCCGATCCTGTCCAACGTTCTGCTGCGCGCCTCCGGCGCCAATCTGGA
CATGAAGGCGA CCGACCTCGATCTGGAAATCACCGAAGCGACCCCGGCC
ATGGTGGAGCAGGCTGGCGCCACCACCGTACCGGCACACCTG CTTTACG
AAATCGTGCGCAAGCTGCCGGATGGTTCCGAAGTGCTTCTGGCGACCAAC
CCGGACGGCTCCTCCATGACCGT TGCGTCCGGCCGCTCGAAATTCTCGC
TGCAATGCCTGCCGGAAGCGGATTTCCCTGACCTCACCGCCGGCACCTTC
AGCC ACACCTTCAAACTGAAGGCGGCCGATCTGAAGATGCTGATCGACC
GGACGCAGTTTGCGATTTCGACCGAAGAGACGCGT TATTACCTGAACGG
CATTTTCTTCCACACCATCGAAAGCAATGGCGAGCTGAAACTGCGCGCCG
TCGCCACCGACGGTCA CCGCCTTGCGCGTGCTGACGTCGATGCGCCCTC
CGGCTCCGAAGGCATGCCGGGCATCATCATTCCGCGCAAGACCGTCG GT
GAACTGCAGAAGCTGATGGACAATCCGGAACTGGAAGTCACAGTCGAAGT
CTCGGATGCGAAGATCCGCCTGGCCATC GGTTCCGTCGTTCTGACCTCG
AAGCTGATCGACGGCACCTTTCCCGATTATCAGCGCGTCATCCCAACCGG
CAACGACAA GGAAATGCGCGTCGATTGCCAGACCTTCGCCCGGGCAGTG
GACCGTGTTTCGACGATTTCTTCCGAGCGCGGCCGCGCCG TGAAGCTGG
CGCTAACTGACGGCCAGTTGACGCTGACCGTCAACAATCCCGACTCGGGA
AGTGCTACCGAAGAAGTGGCC GTTGGCTACGACAATGATTCGATGGAAA
TCGGCTTCAATGCCAAATATCTCCTCGACATCACGTCGCAGCTCTCCGGC
GA AGATGCGATTTTTCTGCTGGCGGATGCGGGTTCGCCAACACTGGTTC
GCGATACCGCCGGCGACGACGCACTCTATGTTC TGATGCCGATGCGCGT
TTAAAACCGACCGTTTTCTTCAATTTTTCCAGAAACGCCGGTGGATCGCT
TCATCGGCGTTTTT TGATTCGGCGAACAGGTGGCTCTACCCGTAACTGA
ATTTTCTCAGTTACGACATTTTGCCTTGTTTTTGCGCCAAATGGG ATCA
ACAGTACGTAACAATTTTTTGACAATGACCAATACATCCGAGGGGAATCA
TGGCACTCAACCTGAAGCAACGGCTT GAACAAAAATTTGAGGAAGAAAT
CCGCTTTTTCAAAGGTATGGTCAGCCAGCCGAAAAAAGTCGGCGCCATTG
TCCCGAC GGTTCCGTCGTTCTGACCTCGAAGCTGATCGACGGCACCTTT
CCCGATTATCAGCGCGTCATCCCAACCGGCAACGACAA GGAAATGCGCG
TCGATTGCCAGACCTTCGCCCGGGCAGTGGACCGTGTTTCGACGATTTCT
TCCGAGCGCGGCCGCGCCG TGAAGCTGGCGCTAACTGACGGCCAGTTGA
CGCTGACCGTCAACAATCCCGACTCGGGAAGTGCTACCGAAGAAGTGGCC
GTTGGCTACGACAATGATTCGATGGAAATCGGCTTCAATGCCAAATATC
TCCTCGACATCACGTCGCAGCTCTCCGGCGA AGATGCGATTTTTCTGCT
GGCGGATGCGGGTTCGCCAACACTGGTTCGCGATACCGCCGGCGACGACG
CACTCTATGTTC TGATGCCGATGCGCGTTTAAAACCGACCGTTTTCTTC
AATTTTTCCAGAAACGCCGGTGGATCGCTTCATCGGCGTTTTT TGATTC
GGCGAACAGGTGGCTCTACCCGTAACTGAATTTTCTCAGTTACGACATTT
TGCCTTGTTTTTGCGCCAAATGGG ATCAACAGTACGTAACAATTTTTTG
ACAATGACCAATACATCCGAGGGGAATCATGGCACTCAACCTGAAGCAAC
GGCTT
22
Three Layers of Genome Annotation. From Stein,
L. 2001. Nature Reviews genetics 2493-503
23

• Human Genome Project
• 1996 first eukaryotic genome (Saccharomyces
  cerevisiae) sequenced and stored into NCBIs
  Genomes division. (Chromosome size 270KB
  1500KB)
• 1997 Entrez Genomes was able to provide the
  first graphical views of genomic sequence data
  (genetic, physical and cytogenetic maps).
• 2001.2 the working draft of the human genome
  (Lander et al., 2001). Chromosome size 46MB
  246MB.
• NCBI created the first human Map Viewer. UCSCs
  human Genome Browser EBIs Ensembl a system to
  annotate automatically the human genome sequence
  as well as to store and visualize the data.
• Today, each site provides free access not only to
  human genome sequence data, but also other
  assembled genomic sequences.

24

• A simple description of each browser
• The backbone is an assembled genomic sequence.
• UCSC (May 2000) and NCBI assembled BAC sequences
  into longer contigs.
• NCBI assembly now is displayed by all three
  browsers (UCSC stopped producing assemblies in
  Dec 2001).
• The sequence is known by its build number at
  NCBI, and by a date at UCSC. (March 2006, NCBI
  Build 36.1, hg18)
• Each of the three genome browsers provides their
  own annotation of the common assembled sequence.
• Location of genes, both known and predicted.
• Different sources of mRNAs, different alignment
  prediction method.
• ESTs and SNPs, the location of STS makers.
• Homologous sequences from other organisms.

25
Depending on the state of sequencing project,
genomic coordinates along the chromosome may
change dramatically from assembly to assembly.
26
DNA Sequence Databases
Data flow for new submissions updates between
the three databases
27
The RefSeq project (NCBI)

• Providing a reference sequence for each molecule
  in the central dogma (DNA, mRNA, and protein).
• Distinct accession number series (26 format)
• NT_123456 Genomic contig (DNA)
• NM_123456 mRNA
• NP_123456 Protein
• XM_123456 Model mRNA
• XP_123456 Model protein

28
Protein Sequence Database (UniProt)

• Swiss-Prot protein database (early 1980s, Amos
  Bairoch) TrEMBL
• Combining Swiss-Prot, TrEMBL, and PIR-PSD
  together.
• Three major components
• The UniProt Archive (UniParc), into which new
  updated sequences are loaded
• The UniProt Knowledgebase, with the goal of
  providing an expertly curated database.
• The UniProt nonredundant reference database
  (UniRef)

29
The flow of data from primary sources into the
component of databases of the Universal Protein
Resource
Complete coverage of all predicted coding
sequences from human, mouse, and rat.
30
(No Transcript)
31
MeSH Terms NLM's Medical Subject Headings
controlled vocabulary of biomedical terms that is
used to describe the subject of each journal
article in MEDLINE. MeSH contains more than
23,000 terms and is updated annually to reflect
changes in medicine and medical terminology. MeSH
terms are arranged hierarchically by subject
categories with more specific terms arranged
beneath broader terms. PubMed allows you to view
this hierarchy and select terms for searching in
the MeSH Database.
32
Lincoln D. Stein
33
Sequence Databases Beyond NCBI

• NCBI is the center of the sequence universe
• There are many specialized sequence databases
  throughout the world that server specific groups
  in the scientific community.

34
(No Transcript)
35
Biological Data Objects
SNPs Human variants

• Laboratory of Computational Molecular Biology
• College of Life Sciences
• Beijing Normal University

36
Why study genetic variation?
Saccharomyces cerevisiae
Caenorhabditis elegans
Drosophila melanogaster
Takifugu rubripes
Xenopus tropicalis
Gallus gallus
Mus musculus
Pan troglodytes
Homo sapiens
(different population variations)
Population A
Population B
Population C
37
Feuk L. et al., 2006. Nature Reviews Genetics.
785-97
Disease susceptibility ??????
38
Feuk L. et al., 2006. Nature Reviews Genetics.
785-97
??
pericentric-???paracentric-????
??
????(uniparental disomy)??????????,???????????????
???
39
Science 315, 848853 (2007)
40

• With the advent of molecular biology, and DNA
  sequencing in particular, smaller and more
  abundant alterations were observed. Such
  differences include
• Single Nucleotide Polymorphisms (SNPs-??????)
• various repetitive elements(????) that involve
  relatively short DNA sequences, e.g., micro- and
  minisatellites(?/??? )
• small (usually lt1 kb) insertions, deletions,
  inversions and duplications(??).
• It was presumed that these small-scale variants
  constitute most genetic variation for example,
  estimates predict that there are at least 10
  million SNPs within the human population,
  averaging 1 every 300 nucleotides among the 3
  billion nucleotide base pairs that constitute the
  genome of an individual.

41
Genetic(genomic) variation

• Mutations fundamentally are produced by errors in
  DNA replication are the ultimate source of
  genetic variation in a population.
• The frequency of alleles (????)depends on the
  evolutionary forces that have been applied to the
  population (e.g., genetic drift(????), selection,
  migration) and on the age of the mutation itself.
• When the mutant allele exceeds 1 representation
  in a population, we refer to the mutation as a
  polymorphism, with no implied association with
  phenotype.

42

• Single nucleotide polymorphisms (SNPs) a
  polymorphism in which alleles are defined by
  single or few base changes sometimes it also
  contains deletions and insertions of one or few
  bases.
• Short tandem repeats (STRs??????? ) a number of
  repeats of short, 2- to 5-nt subsequences.
  Typically, they are scattered throughout the
  genome and are flanked by unique sequences that
  can be used to target a PCR amplification of the
  locus. Most STRs are variable in copy number,
  making them valuable genetic markers.
• Seq1 GTCACTGACACACACAGTACGT
• Seq2 GAC CTGACACACACACACACACACAGTACG

43

• SNP Discovery methods
• Pairwise sequence comparison aligning (??,??)
  two sequences from different individuals DNA and
  to look for high-quality sequence difference.
• Deep resequencing using PCR resequencing of DNA
  samples to find SNPs in a well-defined region of
  the genome.
• Separation method Slab gel, capillary
  electrophoresis (CE?????), MS, DHPLC (Denaturing
  high performance liquid chromatography
  ????????????), etc
• Homogeneous assay Taqman, Invader, etc
• Solid-phase assay beads and DNA arrays chips
  etc.

44
(No Transcript)
45

• NCBI dbSNP content
• Original submitter-supplied data (ss)
• Integrated postsubmission content
• refSNP clusters (rs)
• Mapping results
• Functional analysis
• Average levels of diversity

46
Alleles define the class of the variation

• SNP
• in del DIPs, deletion represented by -
• het variation has unknown sequence composition,
  but is observed to be heterozygous.
• microsatllite microsatellite / simple sequence
  repeat / STR
• mixed MNP multiple nucleotide polymorphism (all
  alleles same length, where length gt 1).
• named allele sequences defined by name tag,
  e.g., (Alu)/-
• no variation invariant region in surveyed
  sequence.
• The dbSNP database is released to the public in
  periodic builds that are synchronized with
  current genome assemblies.

47
Resource Integration

• The clustered data from dbSNP provides a
  non-redundant set of variation from each organism
  in the database.
• The non-redundant set of variations (refSNP
  cluster set) is annotated on reference genome
  sequence contigs, chromosome, mRNAs, and proteins
  as part of the NCBI RefSeq project.
• Summary properties are recomputed for each refSNP
  cluster and are used to build fresh indices for
  NCBIs Entrez query system and to update
  variation maps or tracks in genome browsers
  (e.g., Ensembl, UCSC Genome Browser).

48
Variation Function Class

• Defined by the position of the variation relative
  to the structure of the aligned transcript (may
  having several different functional relationships
  to a gene may having multiple, potentially
  different relationships to its local gene
  neighbors).
• locus region near a gene
• mRNA UTR (Untranslated Region-???????????)
• intron
• splice site
• coding region
• synonymous (????)
• nonsynonymous(?????)

49
Genotyping (????)

• The process of determining the allele states of
  selected polymorphisms in selected groups of
  individuals.
• How to genotype a known STR?
• PCR primers are designed on either side of the
  STR.
• The amplified product from a DNA sample is
  sized-fractionated (??) by electrophoresis to
  determine the number of copies present.
• An individual who is homozygous for one of the
  allele states will show a single predominant
  band those who are heterozygous will show two
  distinct bands.
• For STR polymorphisms, there can be many
  different copy numbers present in the population,
  so the copy number(s) is important to record for
  each sample.

50
(No Transcript)
51

• Uneven distribution
• The distribution of SNPs departs significantly
  from what would be predicted under the standard
  population genetics models gt SNP hot spots
  cold spots.
• What combination of historical, structural or
  selective pressures are responsible for this
  phenomenon?
• How is it related to other factors, such as the
  distribution of genes repetitive elements?

52
Haplotypes(???????) A haplotype is defined as a
specific set of alleles (SNPs) observed on a
region (or the whole) of a given chromosome. When
comparing haplotypes from many individuals,
shared patterns are seen that occur with much
greater likelihood than would be estimated by
assuming each allele state was independent of the
others. This nonrandom association is called
linkage disequilibrium (LD-?????).
53

• How SNPs shape haplotype-structure?
• SNPs in close physical proximity are often
  strongly correlated.
• The length of a shared haplotype segment is
  called a haplotype block.
• There is substantial variation in block sizes
  across the genome.

Haplotype blocks sizable regions over which
there is little evidence for historical
recombination and within which only a few common
haplotypes are observed.
54

• These haplotype blocks are likely to result
  from the fact that recombination, that is, the
  re-shuffling of chromosome segments that occurs
  during formation of sex cells (meiosis-????),
  tends to occur in certain areas of the
  chromosomes more often than in others.
• Thus, any single human chromosome is a mosaic of
  different haplotype blocks, where each block has
  its own pattern of variation. (5kb200kb and 45
  common haplotypes)

Haplotype blocks tend to be shorter in Africa
than elsewhere.
The chromosomes of two hypothetical individuals
are shown. Each individual carries two copies of
each block (as humans carry two sets of
chromosomes).
55

• Usually, Block regions have only a few common
  haplotypes (each
• with a frequency of at least 5), which
  account for most of the
• variation from person to person in population.
• Those common haplotypes could be distinguished
  by a few tag
• SNPs, which represent the most of the
  information on the pattern
• of genetic variation in block region.

Three haplotypes are shown. The two SNPs in color
are sufficient to identify (tag) each of the
three haplotypes.
Theoretically, researchers could look for these
regions by genotyping 10 million SNPs. However,
the methods to do this are currently too
expensive. The HapMap(????) will identify which
200,000 to 1 million tag SNPs provide almost as
much mapping information as the 10 million SNPs.
56
About the International HapMap Project
????????????? Five countries spent 138 million
in total

• Launched in October 2002.
• The goal of the International HapMap Project is
  to develop a public haplotype map of the human
  genome, the HapMap, which will describe the
  common patterns of human DNA sequence variation.
• The HapMap is expected to be a key resource for
  researchers to use to find genes affecting
  health, disease, and responses to drugs and
  environmental factors.

57
???????????? (??YRI) ???????? (??JPT)
???????? (??CHB) CEPH??(???????????????)
(??CEU)
HapMap data release 21, July 2006, on NCBI B35
assembly, dbSNP b125
58
How did HapMap project build ?

• Populations and samples
• The DNA samples for the HapMap will come from a
  total of 270 people
• the Yoruba people in Ibadan, Nigeria (30
  both-parent-and-adult-child trios),
• Japanese in Tokyo (45 unrelated individuals),
• Han Chinese in Beijing (45 unrelated individuals)
• the CEPH (30 trios).
• These numbers of samples will allow the Project
  to find almost all
• haplotypes with frequencies of 5 or higher.
  
• All of the new samples collected for the Project
  are being obtained with
• protocols approved by the appropriate
  ethics (????) committees, after culturally
  appropriate processes of community engagement or
  public consultation and individual informed
  consent.

59

• Genotype data
• The project produce the genotypes of the 270
  individual samples and
• the frequencies of SNP alleles.

Affymetrix
Illumina
60

• Haplotype phasing
• Inferring the haplotype data from
  genotypes data .

Statistics model
Genotype data
Haplotype data

• Haplotype-block partition
• The Project will use standard measures of
  SNP association, such as D'
• and r2, partitioning the ENCODE region
  into block structure.

61
HapMap genotype data dump
62
HapMap file format (genotype)
63
HapMap genotype frequency data dump

• refhom-gt refhom-freq refhom-count
• het-gt het-freq het-count
• otherhom-gt otherhom-freq otherhom-count
• totalcount

64
HapMap allele frequency data dump

• ref_allele ref-allele_freq ref-allele_count
• other_allele other-freq other-allele_count
• totalcount

65
Phased haplotypes
66
(No Transcript)
67
Haplotype-block partition The Project will use
standard measures of SNP association, such as D'
and r2, partitioning a region into block
structure.
As soon as these blocks are identified, only 34
SNPs within a block will be necessary to
characterize the haplotype of that entire segment
of the genome. These are called haplotype tag
SNPs. Across the entire genome, which may contain
up to 10 million common SNPs, it is anticipated
that only 500,000 tag SNPs will be necessary to
identify more than 90 of the haplotype blocks
(1/20), greatly reducing the costs of these
studies.
68
(No Transcript)
69
(No Transcript)
70
(No Transcript)
71
36 adjacent SNPs in an ENCODE region (ENr131.2q37)
D 1 for all markers. In contrast, r 2 values
display a complex pattern, varying from 0.0003 to
1.0., with no relationship to physical
distance. Why?
Only seven haplotypes are observed (five seen
more than once) among the 120 parental CEU
chromosomes studied, reflecting shared ancestry
since their most recent common ancestor among
apparently unrelated individuals.
Simple LD pattern for the region absence of
recombination
72
Selection of tag SNPs for association studies

• We refer to the set of SNPs genotyped in a
  disease study as tags. A given set of tags can be
  analysed for association with a phenotype using a
  variety of statistical methods which we term
  tests, based either on the genotypes of single
  SNPs or combinations of multiple SNPs.
• The shared goal of all tag selection methods is
  to exploit redundancy among SNPs, maximizing
  efficiency in the laboratory while minimizing
  loss of information (this literature is extensive
  and varied, despite its youth).
• Pairwise algorithm SNPs are selected for
  genotying until all common SNPs are highly
  correlated (r2 gt 0.8)

73
(2). Software for Haplotype block partition and
Tag SNPs analysis
Resources for haplotype block analysis
(1). Software for Haplotype Inference

• PHASE (http//www.stat.washington.edu/stephens/sof
  tware.html)
• Haplotyper (http//www.people.fas.harvard.edu/jun
  liu/Haplo/docMain.htm)
• Haplore (http//bioinformatics.med.yale.edu/softwa
  relist.html)

• HapBlock (http//hto-b.usc.edu/msms/HapBlock/)
• HaploBlockFinder (http//cgi.uc.edu/cgi-bin/kzhang
  /haploBlockFinder.cgi)
• TagSNPs (http//www-rcf.usc.edu/
  stram/tagSNPs.html)
• Tagger (http//www.broad.mit.edu/mpg/tagger/ )

(3). Haplotype Visualization Software

• HaploView (http//www.broad.mit.edu/personal/jcbar
  ret/haploview/)
• Haplot (http//info.med.yale.edu/genetics/kkidd/pr
  ograms.html)

74
Association Studies

• A group of individuals are selected as cases
  and another as controls.
• The cases group are individuals that are
  diagnosed with some disease, react to some type
  of medicine, or are even especially healthy.
• The controls group are individuals that do not
  exhibit the feature selected for the cases group.
• For case-control studies, a selection of SNPs is
  genotyped in both the case and control groups,
  and those alleles that exhibit a higher incidence
  (???,???) in the case group as opposed to the
  control group are potential makers for the
  observed phenotype.

75
(No Transcript)
76
NATURE. Vol 445828. 2007
It is the largest GWA study so far, and tackles a
very common disease that is rising in prevalence
throughout the world.
The efforts to understand the interplay between
genetic and environmental risk factors in
generating the high frequency of the disease.
77
(No Transcript)
78
Other Applications

• Evolution studies the map may be biased because
  it refers common SNPs to rare ones.
• Examining genomes architecture
• Nature selection studies

79
Debate

• Common mutations are behind most common diseases?
  (common diseases common mutations)
• OR
• Common diseases arise from combinations of rare
  mutations?

80
Major Analysis Tools

• Laboratory of Computational Molecular Biology
• College of Life Sciences
• Beijing Normal University

81
Outline

• BLAST Sequence Similarity Comparison
• PSI-BLAST Distant Similarity Comparison
• ClustalW Multiple Sequence Alignment
• HMMER More Sensitive Multiple Sequence
  Alignment
• PhylipP Inferring Phylogenies

Comparison of protein and DNA sequence is one of
the foundations of bioinformatics.
82

• Evolution
• The theory of evolution is the foundation upon
  which all of modern biology is built.
• From anatomy to behavior to genomics, the
  scientific method requires an appreciation of
  changes in organisms over time.
• It is impossible to evaluate relationships among
  gene sequences without taking into consideration
  the way these sequences have been modified over
  time.
• Relationships
• Similarity searches and multiple alignments of
  sequences naturally lead to the question
• How are these sequences related?
• And more generally
• How are the organisms from which these sequences
  come related?

83
BLAST
Basic Local Alignment Search Tool (BLAST) is the
tool most frequently used for calculating
sequence similarity.
84
Why BLAST?

• Identify unknown sequences
• Help gene/protein function and structure
  prediction genes with similar sequences tend to
  share similar functions or structure.
• Identify protein family group related (paralog
  or ortholog) genes and their proteins into a
  family.
• Prepare sequences for multiple alignments
• And more

85
What is BLAST?

• An Example of Sequence Comparison

86
Components of Sequence Alignment

• Scoring function a measure of similarity between
  elements (nucleotides, amino acids, gaps)
• An algorithm for alignment
• Confidence assessment of alignment result.

87
Global vs. Local Alignment

• Global Alignment the alignment of complete
  sequences
• Good for comparing members of same protein family
• Needleman Wunsch 1970. J Mol Biol 48443
• Local alignment the alignment of segments of
  sequences
• ignore areas that show little similarity
• Smith Waterman 1981,.J Mol Biol, 147195
• modified from Needelman-Wunsh algorithm
• can be done with heuristics (FASTA and BLAST)

88
BLAST Stages

• Stage I
• Find matching word pairs
• Extend word pairs as much as possible,i.e., as
  long as the total weight increases
• Result High-scoring Segment Pairs (HSPs)
• THEFIRSTLINIHAVEADREAMESIRPATRICKREAD
• INVIEIAMDEADMEATTNAMHEWASNINETEEN
• Stage II
• Try to connect HSPs by aligning the sequences in
  between them
• THEFIRSTLINIHAVEADREA____M_ESIRPATRICKREAD
• INVIEIAMDEADMEATTNAMHEW___ASNINETEEN

89
BLAST Options

• Composition-based statistics (Yes)
• Sequence Complexity Filter (Yes)
• Expect (E) value (10)
• Word Size (3 or 11)
• Substitution or Scoring Matrix (Blosum62)
• Gap Insertion Penalty (11)
• Gap Extension Penalty (1)

90

• BLAST is a collection of programs with versions
  for query-to-database pairs such as
• Query nucleotide ? nucleotide DB blastn
• Query protein ? protein DB lt- nucleotide DB
  tblastn
• Query protein ? protein DB blastp
• Query nucleotide -gt query proteins ? protein DB
  blastx
• Query nucleotide -gt query protein ? protein DB lt-
  nucleotide DB tblastx

91
(No Transcript)
92
BLAST????????
????
???????
??????
?
?
?
?
???????
?
?
??????
?
?
blastp
tblastn
blastn
tblastx
blastx
93
BLAST Parameters

• Identities - No. exact residue matches
• Positives - No. and similar ID matches
• Gaps - No. gaps introduced
• Score - Summed HSP score (S)
• Bit Score - a normalized score (S)
• Expect (E) - Expected of chance HSP aligns
• P - Probability of getting a score gt X
• T - Minimum word or k-tuple score (Threshold)

94
E-value

• The probability that a variate would assume a
  value greater than or equal to the observed value
  strictly by chance P(zgtzo)
• If the E-value found for an alignment is low
  (lt0.001) then alignment is probably biologically
  meaningful.
• Pre-compute the parameters based on a
  statistical model

95
Low complexity issue

• Watch out for
• transmembrane or signal peptide regions
• coil-coil regions
• short amino acid repeats (collagen, elastin)
• homopolymeric repeats
• BLAST uses SEG to mask amino acids
• BLAST uses DUST to mask bases

96
BLAST-related tools

• BLAST2Sequences to find local alignments between
  any two protein or nucleotide sequences.
• MegaBLAST a variation of BLASTN that has been
  optimized specifically for use in aligning either
  long or high similar (gt95) sequences and is the
  method of choice when looking for exact matches
  in nucleotide databases.
• PSI-BLAST (position-specific-iterated BLAST)
  particularly well suited for identifying
  distantly related proteins that may not have been
  found using the traditional BLASTP method.
• BLAT (BLAST-Like Alignment Tool) similar to
  MegaBLAST in that it is designed rapidly to align
  longer nucleotide sequences having more than 95
  similarity but using a slightly different
  strategy than does BLAST to achieve faster speed.

97
PSI-BLAST
Position-Specific Iterated (PSI)-BLAST is the
most sensitive BLAST program, making it useful
for finding very distantly related proteins
98
Distant similarity detection

• Many functionally and evolutionarily important
  protein similarities are recognizable only
  through 3D structures comparison.
• When such structures are not available, patterns
  of conservation identified from the alignment of
  related sequences can aid the recognition of
  distant similarities.
• These conserved patterns variously called
  motifs, profiles, position-specific score
  matrices, and Hidden Markov Models.
• In essence, for each position in the derived
  pattern, every amino acid is assigned a score.
• Highly conserved residue at a particular position
  is assigned a high positive score, and others are
  assigned high negative scores.
• At weakly conserved positions, all residues
  receive scores near zero.
• Position-specific scores can also be assigned to
  potential insertions and deletions.

99

• The power of profile methods can be further
  enhanced through iteration of the search
  procedure .
• Position-specific scores improve the ability of
  successive BLAST iterations for detecting remote
  homologs
• Use PSI-BLAST when your standard protein-protein
  BLAST search either failed to find significant
  hits, or returned hits with descriptions such as
  "hypothetical protein" or "similar to..."

100
PSI-BLAST flow chart
Take a sequence
Search for similar sequences in a full sequence
database
FGLGRT-I-T-YMTN -GLVRT-I---LGLE FGLLRT-I---YMTQ
Sequences are multiply aligned

• After several iterations of this procedure we
  have
• Sequence information, inc. links to annotation
• Several sets of multiple alignments.
• Profiles, derived by us or by PSI-BLAST
• Thresholding information (alignment statistics)

Construct a profile, and represent conservation
in each position numerically
Profile holds more information than a single
sequence use the profile to retrieve additional
sequences
101
(No Transcript)
102
Multiple Alignment Methods

• The most practical and widely used method in
  multiple sequence alignment is the hierarchical
  extensions of pairwise alignment methods.
• The principal is that multiple alignments is
  achieved by successive application of pairwise
  methods.
• Different algorithms for multiple alignment
• CarilloLipman (MSA, DCA)
• Segment based (Dialign)
• Iterative (Profiles, HMMs)
• Progressive (ClustalW, T-Coffee)
• PSI-BLAST and iSCANPS

103
What is a multiple alignment?

• An alignment that contains more than two
  sequences

VTISCTGSESNIGAG-NHVKWYQQLPG VTISCTGTESNIGS--ITVNWY
QQLPG LRLSCSSSDFIFSS--YAMYWVRQAPG LSLTCTVSETSFDD--
YYSTWVRQPPG PEVTCVVVDVSHEDPQVKFNWYVDG-- ATLVCLISDF
YPGA--VTVAWKADS-- AALGCLVKDYFPEP--VTVSWNSG--- VSLT
CLVKEFYPSD--IAVEWWSNG--
104
Why it is important to accurately assess multiple
alignments

• Natural extension of Pairwise Sequence Alignment
• Pairwise alignment whispers multiple
  alignment shouts out loud Hubbard et al 1996
• Much more sensitive in detecting sequence
  relationship and patterns
• Help prediction of the secondary and tertiary
  structures of new sequences
• Preliminary step in molecular evolution analysis
  using Phylogenetic methods for constructing
  phylogenetic trees.
• In order to characterize protein families,
  identify shared regions of homology in a multiple
  sequence alignment
• This happens generally when a sequence search
  revealed homologies to several sequences.
• Identify primers and probes to search for
  homologous sequences in other organisms

105
ClustalWfor multiple alignment

• ClustaW is a general purpose multiple alignment
  program for DNA or proteins.
• ClustalW is produced by Julie D. Thompson, Toby
  Gibson of European Molecular Biology Laboratory,
  Germany and Desmond Higgins of European
  Bioinformatics Institute, Cambridge, UK.
  Algorithmic
• Thompson, J. D., D. G. Higgins, et al. (1994).
  "Clustal-W - Improving the Sensitivity of
  Progressive Multiple Sequence Alignment through
  Sequence Weighting, Position-Specific Gap
  Penalties and Weight Matrix Choice." Nucleic
  Acids Research 22(22) 4673-4680.
• ClustalW can create multiple alignments,
  manipulate existing alignments, do profile
  analysis and create phylogentic trees.
• Fast and reliable are the progressive algorithms
  found in ClustalW.

106
Tips for ClustalW

• Provides global multiple sequence alignment.
• Not constructed to perform local alignments.
• No guarantee to find best alignment.
• No scoring system.
• The very first sequences to be aligned are the
  most closely related in the tree
• if they align well, there will be few errors,
• the more distantly related the more errors.
• Choice of suitable scoring matrices and gap
  penalties

107
When to use?

• for more closely related sequences and large
  number of sequences.
• repeatedly realigns subgroups of sequences then
  aligning these subgroups into global alignment of
  all the sequences, aim is to improve the overall
  alignment score.
• selection of groups is based on the phylogenetic
  tree (separation of one or two sequences from the
  rest) similar to that of progressive alignment.

108
(No Transcript)
109
HMMER
HMMER is a freely distributable implementation of
profile Hidden Markov Models (HMMs) software for
protein sequence analysis.
110

• Hidden Markov Models (HMMs)
• HMMs are statistical models of the primary
  structure consensus of a sequence family.
• HMM- or profile-based methods typically
  outperform pairwise methods in both alignment
  accuracy and database search sensitivity and
  specificity.
• The advantage of HMMs is that HMMs have a formal
  probabilistic basis.

111
Why ?

• Weight matrices do not deal with insertions and
  deletions.
• In alignments, this is done in an ad-hoc manner
  by optimization of the two gap penalties for
  first gap and gap extension.
• HMM is a natural frame work where
  insertions/deletions are dealt with explicitly.
• Alignments can be used to construct hidden Markov
  models
• These are basically statistical models of residue
  preferences gap insertion deletion penalties
  for a specific protein domain
• The theory behind profile HMMs R. Durbin, S.
  Eddy, A. Krogh, and G. Mitchison, Biological
  sequence analysis probabilistic models of
  proteins and nucleic acids, Cambridge University
  Press, 1998.
• HMMER makes this easy.

112
HMMs are trained from a multiple alignment
113
(No Transcript)
114
PHYLIP
A Package of programs for inferring phylogenies
(evolutionary tree)
115
Phylogenetics

• Evolutionary theory states that groups of similar
  organisms are descended from a common ancestor.
• Phylogenetic systematics (cladistics) is a method
  of taxonomic classification based on their
  evolutionary history.
• It was developed by Willi Hennig, a German
  entomologist, in 1950.
• Three major reason to use phylogenetics
• Understand the lineage of different species
• Organizing principle to sort species into a
  taxonomy
• Understand how various functions evolved
• Understand forces and constraints on evolution
• Perform multiple sequence alignment
• Predict gene function (phylogenetic footprint)

116
Species/Gene Trees

• Species tree (how are my species related?)
• contains only one representative from each
  species
• when did speciation take place?
• all nodes indicate speciation events
• Gene tree (how are my genes related?)
• often contains a number of genes from a single
  species
• nodes relate either to speciation or gene
  duplication events
• Your sequence data may not have the same
  phylogenetic history as the species from which
  they were isolated
• Different genes evolve at different speeds, and
  there is always the possibility of horizontal
  gene transfer (HGT).

117
Species tree
118
Gene tree
119
Using DNA or protein sequences

• gt70 identity, use DNA (take more times).
• lt70 identity, use coding from protein (if
  possible) and tread the DNA sequences back onto
  the protein alignment.
• DNA is easier and more accurate, however protein
  is reasonably good.

120
What is Phylip?

• A Package of programs for inferring phylogenies
  (evolutionary tree)
• PHYLIP is the most widely-distributed phylogeny
  package, and competes with PAUP to be the one
  responsible for the largest number of published
  trees.
• It is written and distributed by Joe Felsenstein
  and collaborators (some of the following is
  copied from the PHYLIP homepage).
• Available free over the internet for
  Windows95/98/3.x/NT, DOS, PowerMac and 68k
  Macintosh systems.
• PHYLIP has been in distribution since 1980, and
  has over 15,000 registered users.
• Methods are available in the package include
  parsimony, distance matrix, and likelihood
  methods.
• Also include bootstrapping and consensus tree

121
Phylip input

• Multiple alignment in Phylip format
• Carry out multiple alignment using clustalw
• Save the output in phylip format (.phy)
• Use .phy as an input to phylip

122
Programs

• Seqboot
• Generate a bootstrap set of resample alignments
  for confidence determination.
• PROTPARS
• Estimates phylogenies from protein sequences
  (input using the standard one-letter code for
  amino acids) using the parsimony method, in a
  variant which counts only those nucleotide
  changes that change the amino acid, on the
  assumption that silent changes are more easily
  accomplished.
• DNAPARS
• Estimates phylogenies by the parsimony method
  using nucleic acid sequences. Allows use the full
  IUB ambiguity codes, and estimates ancestral
  nucleotide states. Gaps treated as a fifth
  nucleotide state. It can also fo transversion
  parsimony. Can cope with multifurcations,
  reconstruct ancestral states, use 0/1 character
  weights, and infer branch lengths.
• NEIGHBOR
• Estimate phylogenies from a distance matrix by
  the Neighbor-joining method or UPGMA (Average
  Linkage Clustering) method.
• CONSENSE
• Compute consensus tree from bootstrapped tree
  data.
• See details and other programs
• http//evolution.genetics.washington.edu/phylip/do
  c/main.htmlprograms

123
Steps

• Bootstrap the sequence data
• Generate Phylogenetic trees using Parsimony/ML/NJ
• Consensus tree generation
• Plot the tree

124
(No Transcript)