CS273A
Lecture 4: NON-protein coding genes (ncRNA)
MW 12:50-2:05pm in Beckman B302
Profs: Serafim Batzoglou & Gill Bejerano
TAs: Harendra Guturu & Panos Achlioptas
http://cs273a.stanford.edu [BejeranoFall13/14]
1
Announcements
• HW1 will be out by Friday.
• We’ll announce it over the mailing list/Piazza.
http://cs273a.stanford.edu [BejeranoFall13/14]
2
ATATTGAATTTTCAAAAATTCTTACTTTTTTTTTGGATGGACGCAAAGAAGTTTAATAATCATATTACATGGCATTACCACCATATA
TATCCATATCTAATCTTACTTATATGTTGTGGAAATGTAAAGAGCCCCATTATCTTAGCCTAAAAAAACCTTCTCTTTGGAACTTTC
TAATACGCTTAACTGCTCATTGCTATATTGAAGTACGGATTAGAAGCCGCCGAGCGGGCGACAGCCCTCCGACGGAAGACTCTCCTC
TGCGTCCTCGTCTTCACCGGTCGCGTTCCTGAAACGCAGATGTGCCTCGCGCCGCACTGCTCCGAACAATAAAGATTCTACAATACT
CTTTTATGGTTATGAAGAGGAAAAATTGGCAGTAACCTGGCCCCACAAACCTTCAAATTAACGAATCAAATTAACAACCATAGGATG
AATGCGATTAGTTTTTTAGCCTTATTTCTGGGGTAATTAATCAGCGAAGCGATGATTTTTGATCTATTAACAGATATATAAATGGAA
GCTGCATAACCACTTTAACTAATACTTTCAACATTTTCAGTTTGTATTACTTCTTATTCAAATGTCATAAAAGTATCAACAAAAAAT
TTAATATACCTCTATACTTTAACGTCAAGGAGAAAAAACTATAATGACTAAATCTCATTCAGAAGAAGTGATTGTACCTGAGTTCAA
CTAGCGCAAAGGAATTACCAAGACCATTGGCCGAAAAGTGCCCGAGCATAATTAAGAAATTTATAAGCGCTTATGATGCTAAACCGG
TTTGTTGCTAGATCGCCTGGTAGAGTCAATCTAATTGGTGAACATATTGATTATTGTGACTTCTCGGTTTTACCTTTAGCTATTGAT
TGATATGCTTTGCGCCGTCAAAGTTTTGAACGATGAGATTTCAAGTCTTAAAGCTATATCAGAGGGCTAAGCATGTGTATTCTGAAT
TTAAGAGTCTTGAAGGCTGTGAAATTAATGACTACAGCGAGCTTTACTGCCGACGAAGACTTTTTCAAGCAATTTGGTGCCTTGATG
CGAGTCTCAAGCTTCTTGCGATAAACTTTACGAATGTTCTTGTCCAGAGATTGACAAAATTTGTTCCATTGCTTTGTCAAATGGATC
ATGGTTCCCGTTTGACCGGAGCTGGCTGGGGTGGTTGTACTGTTCACTTGGTTCCAGGGGGCCCAAATGGCAACATAGAAAAGGTAA
GAAGCCCTTGCCAATGAGTTCTACAAGGTCAAGTACCCTAAGATCACTGATGCTGAGCTAGAAAATGCTATCATCGTCTCTAAACCA
ATTGGGCAGCTGTCTATATGAATTAGTCAAGTATACTTCTTTTTTTTACTTTGTTCAGAACAACTTCTCATTTTTTTCTACTCATAA
TTAGCATCACAAAATACGCAATAATAACGAGTAGTAACACTTTTATAGTTCATACATGCTTCAACTACTTAATAAATGATTGTATGA
ATGTTTTCAATGTAAGAGATTTCGATTATCCACAAACTTTAAAACACAGGGACAAAATTCTTGATATGCTTTCAACCGCTGCGTTTT
ATACCTATTCTTGACATGATATGACTACCATTTTGTTATTGTACGTGGGGCAGTTGACGTCTTATCATATGTCAAAGTTGCGAAGTT
TGGCAAGTTGCCAACTGACGAGATGCAGTAACACTTTTATAGTTCATACATGCTTCAACTACTTAATAAATGATTGTATGATAATGT
TCAATGTAAGAGATTTCGATTATCCACAAACTTTAAAACACAGGGACAAAATTCTTGATATGCTTTCAACCGCTGCGTTTTGGATAC
ATTCTTGACATGATATGACTACCATTTTGTTATTGTACGTGGGGCAGTTGACGTCTTATCATATGTCAAAGTCATTTGCGAAGTTCT
GCAAGTTGCCAACTGACGAGATGCAGTTTCCTACGCATAATAAGAATAGGAGGGAATATCAAGCCAGACAATCTATCATTACATTTA
CGGCTCTTCAAAAAGATTGAACTCTCGCCAACTTATGGAATCTTCCAATGAGACCTTTGCGCCAAATAATGTGGATTTGGAAAAAGA
ATAAGTCATCTCAGAGTAATATAACTACCGAAGTTTATGAGGCATCGAGCTTTGAAGAAAAAGTAAGCTCAGAAAAACCTCAATACA
TCATTCTGGAAGAAAATCTATTATGAATATGTGGTCGTTGACAAATCAATCTTGGGTGTTTCTATTCTGGATTCATTTATGTACAAC
GGACTTGAAGCCCGTCGAAAAAGAAAGGCGGGTTTGGTCCTGGTACAATTATTGTTACTTCTGGCTTGCTGAATGTTTCAATATCAA
CTTGGCAAATTGCAGCTACAGGTCTACAACTGGGTCTAAATTGGTGGCAGTGTTGGATAACAATTTGGATTGGGTACGGTTTCGTTG
GCTTTTGTTGTTTTGGCCTCTAGAGTTGGATCTGCTTATCATTTGTCATTCCCTATATCATCTAGAGCATCATTCGGTATTTTCTTC
TTTATGGCCCGTTATTAACAGAGTCGTCATGGCCATCGTTTGGTATAGTGTCCAAGCTTATATTGCGGCAACTCCCGTATCATTAAT
TGAAATCTATCTTTGGAAAAGATTTACAATGATTGTACGTGGGGCAGTTGACGTCTTATCATATGTCAAAGTCATTTGCGAAGTTCT
GCAAGTTGCCAACTGACGAGATGCAGTAACACTTTTATAGTTCATACATGCTTCAACTACTTAATAAATGATTGTATGATAATGTTT
AATGTAAGAGATTTCGATTATCCACAAACTTTAAAACACAGGGACAAAATTCTTGATATGCTTTCAACCGCTGCGTTTTGGATACCT
TCTTGACATGATATGACTACCATTTTGTTATTGTTTATAGTTCATACATGCTTCAACTACTTAATAAATGATTGTATGATAATGTTT
AATGTAAGAGATTTCGATTATCCTTATAGTTCATACATGCTTCAACTACTTAATAAATGATTGTATGATAATGTTTTCAATGTAAGA
TTTCGATTATCCTTATAGTTCATACATGCTTCAACTACTTAATAAATGATTGTATGATAATGTTTTCAATGTAAGAGATTTCGATTA
CTTATAGTTCATACATGCTTCAACTACTTAATAAATGATTGTATGATAATGTTTTCAATGTAAGAGATTTCGATTATCCTTATAGTT
TACATGCTTCAACTACTTAATAAATGATTGTATGATAATGTTTTCAATGTAAGAGATTTCGATTATCCTTATAGTTCATACATGCTT
ACTACTTAATAAATGATTGTATGATAATGTTTTCAATGTAAGAGATTTCGATTATCCTTATAGTTCATACATGCTTCAACTACTTAA
http://cs173.stanford.edu [BejeranoWinter12/13]
3
AATGATTGTATGATAATGTTTTCAATGTAAGAGATTTCGATTATCCTTATAGTTCATACATGCTTCAACTACTTAATAAATGATTGT
“non coding” RNAs (ncRNA)
http://cs273a.stanford.edu [BejeranoFall13/14]
4
Central Dogma of Biology:
5
Active forms of “non coding” RNA
long non-coding
RNA
reverse transcription
(of eg retrogenes)
microRNA
rRNA,
snRNA,
snoRNA
6
What is ncRNA?
• Non-coding RNA (ncRNA) is an RNA that functions without being
translated to a protein.
• Known roles for ncRNAs:
– RNA catalyzes excision/ligation in introns.
– RNA catalyzes the maturation of tRNA.
– RNA catalyzes peptide bond formation.
– RNA is a required subunit in telomerase.
– RNA plays roles in immunity and development (RNAi).
– RNA plays a role in dosage compensation.
– RNA plays a role in carbon storage.
– RNA is a major subunit in the SRP, which is important in protein trafficking.
– RNA guides RNA modification.
– RNA can do so many different functions, it is thought in the beginning there was an
RNA World, where RNA was both the information carrier and active molecule.
7
“non coding” RNAs (ncRNA)
Small structural RNAs (ssRNA)
http://cs273a.stanford.edu [BejeranoFall13/14]
8
ssRNA Folds into Secondary and 3D Structures
AAUUGCGGGAAAGGGGUCAA
CAGCCGUUCAGUACCAAGUC
UCAGGGGAAACUUUGAGAUG
GCCUUGCAAAGGGUAUGGUA
AUAAGCUGACGGACAUGGUC
CUAACCACGCAGCCAAGUCC
UAAGUCAACAGAUCUUCUGU
UGAUAUGGAUGCAGUUCA
Computational
Challenge:
predict 2D & 3D
from sequence.
A
C
A
GA CA
G
C
U
A
G
C
200 G
C
G
U 120
U
P5a CA G
U
G
C
P5 UC G
G
U
A
C
G
C
G
G
G
U
U
AA
A
U
A
A
A
AU
A
A
A
C
180 G
C
G
C
P5c G U A G
C
G
U
A
A
260
A
C
G
A A GG
G
C
C
A
C
P4 C G U
GUUCC G
A
U
A
140
G
U
G
A
U
U U
160 G
C
P6 G
C
G AA
A
U
C
A
C
P5b G
A
U
A
C
U
G
A
G
U
G
220 G
U
C
G
U
A
A
G
C
G P6a
AA
C
G
U
U
A
A
A
G
U
U
A
C
G
A
U
A
U P6b
C
G
A
U
G
C 240
A
U
UCU
Waring & Davies.
(1984) Gene 28: 277.
Cate, et al. (Cech & Doudna).
(1996) Science 273:1678.
9
For example, tRNA
tRNA Activity
ssRNA structure rules
•
•
•
•
Canonical basepairs:
– Watson-Crick basepairs:
• G≡C
• A=U
– Wobble basepair:
• G−U
Stacks: continuous nested
basepairs. (energetically favorable)
Non-basepaired loops:
– Hairpin loop
– Bulge
– Internal loop
– Multiloop
Pseudo-knots
Ab initio RNA structure prediction:
lots of Dynamic Programming
• Objective: Maximizing the number of base pairs
(Nussinov et al, 1978)
simple model:
(i, j) = 1 iff GC|AU|GU
fancier model:
GC > AU > GU
Dynamic programming
 Compute S(i,j) recursively (dynamic
programming)
– Compares a sequence against itself in a dynamic
programming matrix
 Three steps
Initialization
G G G A
A A U C C
G 0
Example:
G 0
0
G
0
A
A
A
U
C
C
GGGAAAUCC
0
0
0
0
0
0
0
0
0
0
0
0
0
the main diagonal
the diagonal below
L: the length of input sequence
Recursion
j
G G G A A A U C C
Fill up the table (DP
matrix) -- diagonal by
diagonal
G 0
0
0
0
G 0
0
0
0
0
G
0
0
0
0
0
0
0
0
0
?
0
0
0
1
0
0
1
1
0
0
0
0
0
0
0
0
0
A
i
A
A
U
C
C
Traceback
G G G A A A U C C
G
0
0
0
0
0
0
1
2
3
G
0
0
0
0
0
0
1
2
3
0
0
0
0
0
1
2
2
0
0
0
0
1
1
1
0
0
0
1
1
1
0
0
1
1
1
0
0
0
0
0
0
0
0
0
G
A
A
A
U
C
C
The structure is:
What are the other “optimal” structures?
Pseudoknots drastically increase
computational complexity
Objective: Minimize Secondary Structure
Free Energy at 37 OC:
Instead of (i, j), measure and sum energies:
-2.1 -0.9 -1.6
C G
G U 
U U
U G 
Ghelix = G G C  + G  C A  + 2G A A  + G A C  =
C G U U U G G G -2.0 kcal/mol - 2.1 kcal/mol + 2x(-0.9) kcal/mol - 1.8 kcal/mol = -7.7 kcal/mol
U
U
G C A A A CA C
G G 
Ghairpin loop = Ginitiation (6 nucleotides) + Gmismatch C A  =


-2.0 -0.9 -1.8 +5.0
5.0 kcal/mol - 1.6 kcal/mol = 3.4 kcal/mol
Gtotal = Ghairpin + Ghelix = 3.4 kcal/mol - 7.7 kcal/mol = -4.3 kcal/mol
Mathews, Disney, Childs, Schroeder, Zuker, & Turner. 2004. PNAS 101: 7287.
http://cs273a.stanford.edu [BejeranoFall13/14]
19
Zuker’s algorithm MFOLD: computing
loop dependent energies
RNA structure
•
•
Base-pairing defines a secondary structure.
The base-pairing is usually non-crossing.
In this example the pseudo-knot makes for an
exception.
Bafna
1
Stochastic context-free grammar
S  aSu
S  aSu
S  cSg
 acSgu
S  gSc
 accSggu
S  uSa
 accuSaggu
Sa
 accuSSaggu
Sc
 accugScSaggu
Sg
 accuggSccSaggu
Su
 accuggaccSaggu
S  SS
 accuggacccSgaggu
 accuggacccuSagaggu
 accuggacccuuagaggu
1. A CFG
2. A derivation of “accuggacccuuagaggu”
3. Corresponding structure
Cool algorithmics. Unfortunately…
–
–
Random DNA (with high GC content) often folds
into low-energy structures.
We will mention powerful newer methods later on.
ssRNA transcription
• ssRNAs like tRNAs are usually encoded by short
“non coding” genes, that transcribe independently.
• Found in both the UCSC “known genes” track, and
as a subtrack of the RepeatMasker track
http://cs273a.stanford.edu [BejeranoFall13/14]
24
“non coding” RNAs (ncRNA)
microRNAs (miRNA/miR)
http://cs273a.stanford.edu [BejeranoFall13/14]
25
MicroRNA (miR)
~70nt
~22nt
miR match to target mRNA
is quite loose.
mRNA
 a single miR can regulate the expression of hundreds of genes.
http://cs273a.stanford.edu [BejeranoFall13/14]
26
MicroRNA Transcription
mRNA
http://cs273a.stanford.edu [BejeranoFall13/14]
27
MicroRNA Transcription
mRNA
http://cs273a.stanford.edu [BejeranoFall13/14]
28
MicroRNA (miR)
Computational challenges:
Predict miRs.
Predict miR targets.
~70nt
~22nt
miR match to target mRNA
is quite loose.
mRNA
 a single miR can regulate the expression of hundreds of genes.
http://cs273a.stanford.edu [BejeranoFall13/14]
29
MicroRNA Therapeutics
Idea: bolster/inhibit miR production to
broadly modulate protein production
Hope: “right” the good guys and/or
“wrong” the bad guys
Challenge: and not vice versa.
~70nt
~22nt
miR match to target mRNA
is quite loose.
mRNA
 a single miR can regulate the expression of hundreds of genes.
http://cs273a.stanford.edu [BejeranoFall13/14]
30
Other Non Coding Transcripts
http://cs273a.stanford.edu [BejeranoFall13/14]
31
lncRNAs (long non coding RNAs)
Don’t seem to fold into clear structures (or only a sub-region does).
Diverse roles only now starting to be understood.
Example: bind splice site, cause intron retention.
 Hard to detect or predict function computationally (currently)
http://cs273a.stanford.edu [BejeranoFall13/14]
32
lncRNAs come in many flavors
http://cs273a.stanford.edu [BejeranoFall13/14]
33
X chromosome inactivation in mammals
X
X
X
Dosage compensation
X
Y
Xist – X inactive-specific transcript
Avner and Heard, Nat. Rev. Genetics 2001 2(1):59-67
Transcripts, transcripts everywhere
Human Genome
Transcribed* (Tx)
Tx from both strands*
* True size of set unknown
http://cs273a.stanford.edu [BejeranoFall13/14]
36
Or are they?
http://cs273a.stanford.edu [BejeranoFall13/14]
37
The million dollar question
Human Genome
Transcribed* (Tx)
Leaky tx?
Tx from both strands*
Functional?
* True size of set unknown
http://cs273a.stanford.edu [BejeranoFall13/14]
38
System output measurements
We can measure non/coding gene expression!
(just remember – it is context dependent)
1. First generation mRNA (cDNA) and EST sequencing:
In UCSC Browser:
http://cs273a.stanford.edu [BejeranoFall13/14]
39
2. Gene Expression Microarrays (“chips”)
http://cs273a.stanford.edu [BejeranoFall13/14]
40
3. RNA-seq
“Next” (2nd) generation sequencing.
http://cs273a.stanford.edu [BejeranoFall13/14]
41
Gene Finding II: technology dependence
Challenge:
“Find the genes, the whole genes, and nothing but the genes”
We started out trying to predict genes directly from the genome.
When you measure gene expression, the challenge changes:
Now you want to build gene models from your observations.
These are both technology dependent challenges.
The hybrid: what we measure is a tiny fraction of the space-time state
space for cells in our body. We want to generalize from measured
states and improve our predictions for the full compendium of states.
http://cs273a.stanford.edu [BejeranoFall13/14]
42
4. Spatial-temporal maps generation
http://cs273a.stanford.edu [BejeranoFall13/14]
43
Cluster all genes for differential expression
Experiment Control
(replicates) (replicates)
genes
Most significantly up-regulated genes
Unchanged genes
Most significantly down-regulated genes
http://cs273a.stanford.edu [BejeranoFall13/14]
44
Determine cut-offs, examine individual genes
Experiment Control
(replicates) (replicates)
genes
Most significantly up-regulated genes
Unchanged genes
Most significantly down-regulated genes
http://cs273a.stanford.edu [BejeranoFall13/14]
45
Genes usually work in groups
Biochemical pathways, signaling pathways, etc.
Asking about the expression perturbation of groups of
genes is both more appealing biologically, and more
powerful statistically (you sum perturbations).
http://cs273a.stanford.edu [BejeranoFall13/14]
46
Ask about whole gene sets
+
Exper. Control
Gene
Set 1
Gene
Set 2
Gene
Set 3
Gene set 3
up regulated
ES/NES statistic
Gene set 2
down regulated
http://cs273a.stanford.edu [BejeranoFall13/14]
The Gene Sets to test come from the Ontologies
TJL-2004
48
One approach: GSEA
Dataset distribution
Number of genes
Gene set 3 distribution
Gene set 1 distribution
Gene Expression Level
http://cs273a.stanford.edu [BejeranoFall13/14]
49
Another popular approach: DAVID
Input: list of genes of interest (without expression values).
http://cs273a.stanford.edu [BejeranoFall13/14]
50
Multiple Testing Correction
run tool
Note that statistically you cannot just run individual tests on 1,000
different gene sets. You have to apply further statistical corrections,
to account for the fact that even in 1,000 random experiments a
handful may come out good by chance.
(eg experiment = Throw a coin 10 times. Ask if it is biased.
If you repeat it 1,000 times, you will eventually get an all heads
series, from a fair coin. Mustn’t deduce that the coin is biased)
http://cs273a.stanford.edu [BejeranoFall13/14]
51
What will you test?
run tool
Also note that this is a very general approach to test gene lists.
Instead of a microarray experiment you can do RNA-seq.
Instead of up/down-regulated genes you can test all the genes in a
personal genome where you see “bad” mutations.
Any gene list can be tested.
http://cs273a.stanford.edu [BejeranoFall13/14]
52
Coding and non-coding gene production
To change its behavior
a cell can change the
repertoire of genes and
ncRNAs it makes.
The cell is constantly
making new proteins and
ncRNAs.
These perform their
function for a while,
And are then degraded.
Newly made coding and
non coding gene products
take their place.
The picture within a cell is
constantly “refreshing”.
http://cs273a.stanford.edu [BejeranoFall13/14]
53
Cell differentiation
To change its behavior
a cell can change the
repertoire of genes and
ncRNAs it makes.
That is exactly what happens
when cells differentiate during
development from stem cells
to their different final fates.
http://cs273a.stanford.edu [BejeranoFall13/14]
54
Human manipulation of cell fate
To change its behavior
a cell can change the
repertoire of genes and
ncRNAs it makes.
We have learned (in a dish) to:
1 control differentiation
2 reverse differentiation
3 hop between different states
http://cs273a.stanford.edu [BejeranoFall13/14]
55
Cell replacement therapies
We want to use this
knowledge to provide a
patient with healthy self
cells of a needed type.
We have learned (in a dish) to:
1 control differentiation
2 reverse differentiation
3 hop between different states
(iPS = induced pluripotent stem cells)
http://cs273a.stanford.edu [BejeranoFall13/14]
56
How does this happen?
Different cells in our body
hold copies of (essentially)
the same genome.
Yet they express very
different repertoires of
proteins and non-coding
RNAs.
How do cells do it?
A: like they do everything else: using their proteins & ncRNAs…
http://cs273a.stanford.edu [BejeranoFall13/14]
57
Gene Regulation
Some proteins and non
coding RNAs go “back”
to bind DNA near
genes, turning these
genes on and off.
Gene
DNA
Proteins
To be continued…
http://cs273a.stanford.edu [BejeranoFall13/14]
58
Inferring Gene Expression Causality
Measuring gene expression over time provides sets of
genes that change their expression in synchrony.
• But who regulates whom?
• Some of the necessary regulators may not change their
expression level when measured, and yet be essential.
“Reading” enhancers can provide gene regulatory logic:
• If present(TF A, TF B, TF C) then turn on nearby gene X
http://cs273a.stanford.edu [BejeranoFall13/14]
59
Review
Lecture 6
• Central dogma recap
– Genes, proteins and non coding RNAs
• RNA world hypothesis
• Small structural RNAs
– Sequence, structure, function
– Structure prediction
– Transcription mode
• MicroRNAs
– Functions
– Modes of transcription
• lncRNAs
– Xist
• Genome wide (and context wide) transcription
– How much?
– To what goals?
• Gene transcription and cell identity
– Cell differentiation
– Human manipulation of cell fates
• Gene regulation control
http://cs273a.stanford.edu [BejeranoFall13/14]
60