RNA secondary structure prediction and gene finding

I519 Introduction to Bioinformatics, 2012
Genome Variations & GWAS
Genome variations
underlie phenotypic
cause inherited
sequence variations can be used for gene
mapping, definition of population structure,
and performance of functional studies.
1000 Genomes Project
 An international collaboration to produce an
extensive public catalog of human genetic
variation, including SNPs and structural
variants, and their haplotype contexts. This
resource will support genome-wide association
studies and other medical research studies.
 The genomes of about 2500 unidentified people
from about 25 populations around the world will be
sequenced using next-generation sequencing
 Results of the pilot phase of the project published
in Nature Volume: 467, Pages: 1061–1073, 2010
How do we find sequence variations?
• look at multiple sequences
from the same genome region
• use base quality values to decide if
mismatches are true polymorphisms
or sequencing errors
Automated polymorphism discovery
PolyBayes (Ref: A general approach to single-nucleotide
polymorphism discovery, Marth et al., Nature Genetics, 1999)
Determine if a genetic difference is due to sequencing error or it is
a real SNP by using base quality values in a rigorous, Bayesian
scheme to compare sequences of arbitrary quality standards.
PYROBAYES (Ref: An improved base-caller for SNP discovery in
pyrosequences. Nature Methods. 2008;5:179-81.)
SNP functional categories
 coding nonsynonymous
– Missense, nonsense, frame shift
 coding synonymous
 Intronic
– splice site
 mRNA utr
– 5' utr or 3' utr
 (gene) locus region (5’ or 3’ to the gene)
– ‘near gene’ usually means within ~2000bp of
 genomic/extragenic (distant from any gene)
SNP nomenclature
 The Human Genome Variation Society
(http://www.hgvs.org/mutnomen/recs.html) has
proposed some guidelines for SNP
nomenclature, but at the moment, there is
minimal consistency.
 Different sources will refer to the same SNP in
different ways
 While dbSNP identifiers (rs#12345678) are
becoming common, they are not required of
publishing authors and not used in all cases.
SNPs at base-pair level
 The base-pair change is given in various forms:
C>T 432G>C
The HGVS nomenclature recommendations:
"c." for a coding DNA sequence (like c.76A>T)
"g." for a genomic sequence (like g.476A>T)
"m." for a mitochondrial sequence (like
"r." for an RNA sequence (like r.76a>u)
 SNP database from NCBI, build 130 contains
63,751,769 refSNP clusters (19,576,037
 dbSNP contains:
– Single nucleotide substitutions
– Small insertion/deletion polymorphism
– Microsatellite repeats
dbSNP content
The SNP database has two major classes of
 Submitted data, i.e., original observations of
sequence variation: Submitted SNPs (SS) with
ss# (e.g, ss5586300)
 Reference Cluster ID: Computed/curated data
(Ref SNP with rs#, e.g., rs25)
 Ref SNP
– Ref SNP Clusters define a non-redundant set of
– Ref SNP clusters may contain multiple submitted
Reference SNP clusters
 Ref SNP clusters are computer-generated and
curated by NCBI staff
 Ref SNP Clusters define a non-redundant set of
 All individual SNPs submitted by a researcher
are given a submitter SNP number (ss#) and
then redundant (repetitive) submitter SNPs are
combined into a RefSNP cluster record, with a
unique rs#
 Ref SNP clusters may contain multiple submitted
An example
Promises of SNPs
 Each person's SNP pattern is unique
 Most SNPs are not responsible for a disease state.
But they can be located near a gene associated with
a certain disease. So SNPs may serve as biological
markers for pinpointing a disease on the human
genome map.
 Application of association study can detect
differences between the SNP patterns of two groups
(control-disease), thereby indicating which pattern is
most likely associated with the disease-causing
 Using SNPs to study the genetics of drug response
will help in the creation of "personalized" medicine.
Annotation of SNPs
 A straightforward and reliable method based on physical
and comparative considerations that estimates the impact
of an amino acid replacement on the three-dimensional
structure and function of the protein (~20% of common
human non-synonymous SNPs predicted to be
deleterious). Ref: Human Molecular Genetics, 2001, Vol.
10, No. 6 591-597
 SIFT: predicting amino acid changes that affect protein
function (used to distinguish between functionally neutral
and deleterious amino acid changes in mutagenesis
studies and on human polymorphisms). Ref: Nucleic
Acids Research, 2003, Vol. 31, No. 13 3812-3814
 Review: Next generation tools for the annotation of
human SNPs by Rachel Karchin, Briefings in
Bioinformatics 2009 10(1):35-52
Genome-wide association study (GWAS)
 A genome-wide association study is an approach that involves
rapidly scanning markers across the complete sets of DNA, or
genomes, of many people to find genetic variations associated
with a particular disease. (http://www.genome.gov/20019523)
 If genetic variations are more frequent in people with the disease,
the variations are said to be "associated" with the disease.
 Genome-wide association study of 14,000 cases of seven
common diseases and 3,000 shared controls (Nature 447, 661678)
 Validating, augmenting and refining genome-wide association
signals (Nature Reviews Genetics 10, 318-329, 2009)
The underlying rationale for GWAS
 'common disease, common variant' hypothesis,
positing that common diseases are attributable in
part to allelic variants present in more than 1–5% of
the population
 SNP genotyping chips – common variants
 most common variants individually or in combination
confer relatively small increments in risk (1.1–1.5fold) and explain only a small proportion of
heritability. E.g. at least 40 loci have been
associated with human height (with an estimated
heritability of about 80%), yet they explain only about
5% of phenotypic variance
Ref: Finding the missing heritability of complex diseases; Nature 461,
747-753, 2009
Explaining missing heritability
 Rare variants;
– variants of low minor allele frequency (MAF; 0.5% < MAF
< 5%), or of rare variants (MAF < 0.5%)
– Rare variants are not sufficiently frequent to be captured by
current GWA genotyping arrays
– And they don’t carry sufficiently large effect sizes to be
detected by classical linkage analysis in family studies
– The primary technology for the detection of rare SNPs is
sequencing, which may target regions of interest, or may
examine the whole genome.
 Structural variation, including copy number variants
(CNVs, such as insertions and deletions) and copy
neutral variation (such as inversions and
Feasibility of identifying genetic variants by risk allele frequency and
strength of genetic effect (odds ratio).
TA Manolio et al. Nature 461, 747-753 (2009) doi:10.1038/nature08494
Genome-wide significance
 Associations that have been identified from a single
GWA data set rarely have definitive statistical
support. p values of <10-7 are required for genomewide significance. A p value of approximately 10-7 in
the GWA setting corresponds to a p value of
approximately 0.05 for a traditional, classical
epidemiological study in which only one hypothesis
is being tested.
 Statistical significance for genomewide studies
– PNAS 100(16):9440-9445, 2003
– q value; similar to the well known p value, except it is
a measure of significance in terms of the false
discovery rate rather than the false positive rate.
Analysis methods
(a) is a baseline analysis; (b)-(e) apply further prior hypotheses
Chi-square statistic tests
O1=a, E1=nAAncase/n, and so on
Population stratification
 Population stratification is the presence of a
systematic difference in allele frequencies
between subpopulations in a population possibly
due to different ancestry.
 Case control association studies assume that
any difference in the SNP genotypes between
the cases and controls is due solely to their
difference in disease status, but not difference in
their genetic background.
 Potential population stratification needs to be
corrected in association studies
GWAS vs genetic linkage method
 Genetic linkage combined with positional cloning
leads to the finding of gene mutations that are
involved with monogenic disease, such as cystic
fibrosis and Huntington's disease. These
mutations most likely alter the amino acid
sequence of protein.
 Most loci that have been discovered through
genome-wide association analysis do not map to
amino acid changes in proteins (with a few
important exceptions). They are predicted to
affect gene expression.
 Bioinformatics challenges for genome-wide
association studies
– Bioinformatics (2010) 26 (4): 445-455
 Finding the missing heritability of complex
– Nature (2009) 461, 747-753

similar documents