Transcript CSC598BIL675-2016

Gene Expression
Expressed genes are those that have been transcribed.
A gene expression profile of a cell is the snapshot of which
genes are expressed in that cell at the time the sample was
taken.
Knowing which genes are expressed in a cell allows:
identification of new genes or transcripts
comparison of expression profiles between samples
Main motives: disease, development, dynamic responses
Differential Gene Expression
Expression profile of genes in one sample vs another.
Different cells, tissues, disease states, developmental stages,
culture conditions, etc, can be compared.
Measure both, subtract the overlap, obtain the difference,
interpret it.
Pay due attention to controls, negative and positive
Pay due attention to range of variability within samples
High throughput and higher throughput
Differential Gene Expression Workflow
1. Formulate the biological questions
2. Experimental design
Which platform, which controls, how many replicates
3. Run the experiment
4. Image processing (by machine)
5. Low-level analysis
Data preprocessing (normalisation step)
5. High-level analysis
Data analysis
6. Reach biological conclusions
Interpretation of results
High Throughput Methods- Advantages
Fast
a lot of data produced quickly
Comprehensive
entire genomes in one experiment
Easy
submit RNA samples to a core facility
Cost
getting cheaper still
High Throughput Methods - Disadvantages
Cost
- Some researchers can’t afford to do
appropriate numbers of controls, replicates
RNA
Significance
- The final product of gene expression is a protein
- How do you filter out non-coding RNA, or transcripts
that are not translated?
Quality
Control
- Artifacts with image analysis and data analysis
- Not enough attention to experimental design
- Need more collaboration with computational scientists
Measuring Differential Gene Expression
• Quantification of mRNA transcripts
EST libraries
SAGE libraries
• Microarray technology
• High throughput RNA-seq technology
trancriptome sequencing
( quantification of mRNA transcripts)
EST Libraries
Expressed Sequence Tags
Single sequencing reads from cDNA libraries• 250 (earlier) - 800 (later) bases long
• usually from 5’ or 3’ end according to cloning strategies
• an indication of which mRNA are in that cell at that time
Highly expressed genes = many ESTs
Low expression genes = fewer (or no detectable) ESTs
Can miss very low level transcripts, and some transcript variants
OK for quantification, known inaccuracies inherent in the method
Not OK for discovery of rare transcripts, too much noise from common
transcripts
dbEST at NCBI
SAGE Libraries
Serial analysis of gene expression
• 14 bp fragment is enough to uniquely
identify a transcript.
• Make cDNA library, cut it to 14bp fragment
per transcript.
• Ligate tags into long concatemers separated
by a marker, and sequence them.
• Output is a quantifiable list of short tags
denoting presence of a gene and how much
of it is there.
Useful in comparing transcriptomes and in
discovery of new genes or transcripts
Microarray Technology
• Single stranded lawn of DNA probes attached
to a membrane on a microarray chip
Usually 20-30 bp of a unique sequence from
a gene. Often have more than one probe
representing a gene.
•
Target (hybridization extract)
Total cDNA extracted from biological sample
and labeled with fluorescent dye
• Targets hybridise to the probes that have
complimentary sequences. Intensity of
hybridization is measured as an indication of
the presence of that gene in the biological
sample.
Uses of microarrays
•
Changes in Gene Expression levels (one or two colours)
Probes are ssDNA (cDNA or oligos),
Target is labeled cDNA derived from mRNA.
•
Genomic Gains and Losses (two colours)
CGH (Comparative Genomic Hybridization)
Probes are ssDNA (oligos)
Target is labeled DNA derived from genomic DNA
• Genomic SNPs (one colour)
Probes are short genomic sequences containing SNPs
Target is labeled DNA derived from genomic DNA
Holdus, Stavrum, Petersen and Stansberg 2008
Image Processing
This is computerized - you just see the final result in a spreadsheet.
The software scans the array and quantitates the signal values, i.e.
converts fluorescence intensity to digital value
Data Preprocessing
You choose the parameters, software does the work
• Background subtraction: Eliminates background noise
• Normalization: This step takes care of
• Unequal quantity of starting sample
• Difference in labeling efficiency
• Difference in detection efficiency
• System biases, etc.
Brings all samples into a similar range of distribution
• Statistical QC
Removes low quality samples and probesets
Detection of Significantly Differentially Expressed Genes
Statistical tests
• Student’s t test for two conditions/groups (control vs treated)
(i.e. the comparison of the means and standard deviations of two
bell shaped curves, based on a t-statistic, testing the nullhypothesis that both distributions came from the same
distribution)
• ANOVA analysis (control vs treatment 1 vs treatment 2)
(i.e. ANalysis Of VAriances: Allows to test the null hypothesis that
the differences within and between at least 3 groups are the same
on average.
Based on F-statistic, the ratio of the variance calculated among the
means to the variance within the samples)
Detection of Significantly Differentially Expressed Genes
• 2-way ANOVA (eg 2 cell lines, 2 treatments)
(i.e. The two-way ANOVA compares the mean differences between
groups that have been split on two independent variables called factors.
The primary purpose of a two-way ANOVA is to understand if there is an
interaction between the two independent variables on the dependent
variable).
All these methods produce p-values to assess the probability to
obtain the result by chance. Problem: What happens if we have
many such tests?
Detection of Significantly Differentially Expressed Genes
Multiple testing problem:
• Say you have a set of hypotheses that you wish to test simultaneously. Let’s,
consider a case where you have 20 hypotheses to test, and a significance level
of a = 0.05. What’s the probability of observing at least one significant result
just due to chance?
P(at least one significant result) = 1 − P(no significant results)
= 1−(1−0.05)20
≈ 0.64
We have a 64% CHANCE to find one significant result randomly …
Detection of Significantly Differentially Expressed Genes
Correction methods:
• Bonferroni (very conservative): significance threshold is a/N
• FDR (False Discovery Rate): check if the kth ordered p-value is
larger than (k × a)/N
• q-value:
chance that p-values in this column
are false positives: q-value
Detection of Significantly Differentially Expressed Genes
Fold change
• Difference in the intensity of a sample vs control or another
sample, indicative of difference in level of expression of the
gene
• Threshold > 2, or > 1.6 in some cases
Then clustering
Then clustering
• In differential gene expression, you are looking for genes that
behave differently between one sample and another, either
up- or down• Once you get your DE gene set, you group the genes
according to similar expression, and the outliers become more
obvious
• Clustering methods similar to those of phylogenetics, but
without the evolutionary weightings, ie distance matrices
More downstream analysis later in the course
RNA-seq
Same concept as sequencing ESTs and counting SAGE tags,
but does not stop at short segments and tags.
What is being sequenced is the cDNA from the mRNA component.
Sequencing of whole transcriptome of a sample (NGS), and
comparing it against the whole transcriptome of another sample.
Costly, informative, bioinformatics not yet fully sorted outwhen does a lot of data become too much data?
Finding the real transcripts
It’s all about the alignment
• First, you align your reads to a reference genome or genomic
region (or assemble the reads de novo)
BWA, Bowtie2, etc
• Then you use a splice-aware aligner, such as TopHat or STAR,
to refine the aligments according to coding sequences (exons)
using known and/or predicted splice junctions
Quantifying reads per gene
• Your aim is to count sequence reads per gene
• When mapping reads to genome:
•
•
•
•
•
Filter out rRNA, tRNA, mitRNA, etc
Filtering out (or in!) non-coding RNA
Deal with alternative splicing
Deal with overlapping genes, pseudogenes
Small reads mean many short overlaps at one end or
the other of intron gaps
• Allele specific gene expression
Some Solutions
1.
Can create a library of transcripts and map reads to transcripts
(still have some ambiguity for multiple isoforms)
[limited, few (if any) use this method]
2.
Can create a library of splice-junctions (span intron gaps)
[Illumina CASAVA uses this method]
3.
Can predict transcripts from genome mapped RNA-seq reads plus
known splice junctions plus predicted splice junctions
[TopHat]
4.
Can do de novo assembly of new transcripts from reads
[Trinity]
c.f. S. Brown, NYU
Normalization
Coverage is not exactly the same for each sample
• Problem: Need to scale RNA counts per gene to total sample
coverage
• Solution – divide counts per million reads
• Problem: Longer genes have more reads, gives better chance
to detect DE
• Solution – divide counts by gene length
• Result = RPKM and later FRKM
(Reads/Fragments Per KB per Million)
c.f. S. Brown, NYU
Better Normalization
• FPKM assumes:
• Total amount of RNA per cell is constant
• Most genes do not change expression
• FPKM is invalid if there are a few very highly expressed genes
that have dramatic change in expression (dominate the pool
of reads)
• Many now use “Quantile” normalization
• New normalization methods currently being published
• Different normalization methods give different results
c.f. S. Brown, NYU
Better Normalization
quantile normalization: making distributions identical in statistical properties
genes
arrays
rearrange
columns
assign ranks
genes
arrays
rank values
assign
values
c.f. S. Brown, NYU
Statistics of Differential Gene Expression
• mRNA levels are variable in cells/tissues/organisms over
time/treatment/tissue etc.
• Need enough replicates to separate biological variability from
experimental variability
• If there is high experimental variability, then variance within
replicates will be high, statistical significance for DE will be
difficult to find.
• Best methods to discover DE are coupled with sophisticated
approaches to normalization
• Very low expressing genes are tricky: FPKM<1
c.f. S. Brown, NYU
Gene Expression Analysis
Databases:
GEO from NCBI
ArrayExpress from EBI
Commercial software:
GeneSpring GX, CLC Bio, many others
Free:
Mostly R based
Not being scared of statistics is an advantage
New methods and algorithms continually being published
Routine experiments are routine, innovative methods more care
The really tricky part is the interpretation of the results
https://github.com/ccsstudentmentors/tutorials/wiki/CCS-Student-Mentors---Tutorials
Suggested additional reading: