10/26/05 promoter prediction (really!)

10/26/05 D Dobbs ISU - BCB 444/544X: Promoter Prediction (really!) 1

10/26/05

Promoter Prediction(really!)


Announcements• BCB Link for Seminar Schedules (updated)

http://www.bcb.iastate.edu/seminars/index.html

Seminar (Fri Oct 28)

12:10 PM BCB Faculty Seminar in E164 Lagomarcino

Assembly and Alignment of Genomic DNA Sequence Xiaoqiu Huang, ComS

http://www.bcb.iastate.edu/courses/BCB691-F2005.html#Oct%2028

Mark your calendars:1:10 PM Nov 14 Baker Seminar in Howe Hall Auditorium

"Discovering transcription factor binding sites"Douglas Brutlag,Dept of Biochemistry & MedicineStanford University School of Medicine


Announcements

BCB 544 Projects - Important Dates:

Nov 2 Wed noon - Project proposals due to David/Drena

Nov 4 Fri 10A - Approvals/responses to students

Dec 2 Fri noon - Written project reports due

Dec 5,7,8,9 class/lab - Oral Presentations (20')

(Dec 15 Thurs = Final Exam)


Announcements

Lab 9 - due Wed noon (today)Exam 2 - this Friday

Posted Online: Exam 2 Study Guide 544 Reading Assignment (2

papers) Lab Keys (today)

Thurs No Lab - Extra Office Hrs instead: David 1-3 PM in 209 Atanasoff Drena 1-3 PM in 106 MBB


Promoter Prediction RNA Structure/Function Prediction

Mon Quite a few more words re: Gene prediction

Wed Promoter prediction

next Mon: RNA structure & function

RNA structure prediction2' & 3' structure prediction

miRNA & target prediction


Optional - but very helpful reading:

1) Zhang MQ (2002) Computational prediction of eukaryotic protein-coding genes. Nat Rev Genet 3:698-709

http://proxy.lib.iastate.edu:2103/nrg/journal/v3/n9/full/nrg890_fs.html

2) Wasserman WW & Sandelin A (2004) Applied bioinformatics for identification of regulatory elements. Nat Rev Genet 5:276-287


03489059922

(that's a hint!)

Check this out: http://www.phylofoot.org/NRG_testcases/


Reading Assignment (for Mon)

Mount Bioinformatics• Chp 8 Prediction of RNA Secondary Structure • pp. 327-355• Ck Errata:

http://www.bioinformaticsonline.org/help/errata2.html

Cates (Online) RNA Secondary Structure Prediction Module• http://cnx.rice.edu/content/m11065/latest/


Review last lecture:

Flowchart for Gene Prediction

Performance Assessment Measures

Correction re: slide 10/24 # 27

Promoters


Gene prediction flowchart

Fig 5.15Baxevanis & Ouellette 2005


Evaluation of Splice Site Prediction


What do measures really mean?

Sp =


Correction re: last lecture:

GeneSeqer Performance Graphs

Brendel et al (2004) Bioinformatics 20: 1157


0.00

0.20

0.40

0.60

0.80

1.00

-10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 200.00

0.20

0.40

0.60

0.80

1.00

-10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20

SnSn

HumanGT site

HumanAG site

0.00

0.20

0.40

0.60

0.80

1.00

-10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20

0.00

0.20

0.40

0.60

0.80

1.00

-10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20

SnSn

A. thalianaAG site

A. thalianaGT site

Brendel 2005

Performance?

Note: these are not ROC curves (plots of (1-Sn) vs Sp)

• But plots such as these (& ROCs) much better than using "single number" to compare different methods• Both types of plots illustrate trade-off: Sn vs Sp


QuickTime™ and aTIFF (LZW) decompressorare needed to see this picture.

Fig 2 - Brendel et al (2004) Bioinformatics 20: 1157


Bayes Factor as Decision Criterion

H0: H=T:}){1(

}{

})|{1(

}|{

Tp

Tp

STp

STpBF

−−=

2-class model: }|{}|{ FSpTSpBF =

7 class model: ∑∑

∑∑

=

=

=

==ix x

ix xx

x x

x xx

Fp

FpFSp

Tp

TpTSpBF

,0,2,1

,0,2,1

0,2,1

0,2,1

}{

}{}|{

}{

}{}|{

Brendel 2005


Evaluation of Splice Site Prediction

• Normalized specificity: αα β

=−

− +1

1

ActualTrue False

PP=TP+FP

PN=FN+TN

AP=TP+FNAN=FP+TN

PredictedTrue

False TNFN

FPTP

Brendel 2005

• Specificity: rAN

AP=

• Misclassification rates: α =FN

APβ =

FP

AN

• Sensitivity: = Coverage


Careful: different definitions for "Specificity"

ActualTrue False

PP=TP+FP

PN=FN+TN

AP=TP+FNAN=FP+TN

PredictedTrue

False TNFN

FPTP

• Specificity:

• Sensitivity:

cf. Guig�ó definitions Sn: Sensitivity = TP/(TP+FN)

Sp: Specificity = TN/(TN+FP) = Sp-

AC: Approximate Coefficient = 0.5 x ((TP/(TP+FN)) + (TP/(TP+FP)) + (TN/(TN+FP)) + (TN/(TN+FN))) - 1

Other measures? Predictive Values, Correlation Coefficient

Brendel definitions


Best measures for comparing different methods?

• ROC curves (Receiver Operating Characteristic?!!)

http://www.anaesthetist.com/mnm/stats/roc/

"The Magnificent ROC" - has fun applets & quotes:

"There is no statistical test, however intuitive and simple, which will not be abused by medical researchers"

• Correlation Coefficient(Matthews correlation coefficient (MCC)

MCC = 1 for a perfect prediction 0 for a completely random assignment

-1 for a "perfectly incorrect" prediction

QuickTime™ and aTIFF (LZW) decompressor

are needed to see this picture.

Do not memorize this!


PromotersWhat signals are there?

Simple ones in prokaryotes

BIOS Scientific Publishers Ltd, 1999

Brown Fig 9.17


Prokaryotic promoters

• RNA polymerase complex recognizes promoter sequences located very close to & on 5’ side (“upstream”) of initiation site

• RNA polymerase complex binds directly to these. with no requirement for “transcription factors”

• Prokaryotic promoter sequences are highly conserved

• -10 region • -35 region


QuickTime™ and aTIFF (LZW) decompressor

are needed to see this picture.

What signals are there? Complex ones in eukaryotes!

Fig 9.13Mount 2004


Simpler view of complex promoters in eukaryotes:



Eukaryotic genes are transcribed by 3 different RNA polymerases


Brown Fig 9.18

Recognize different types of promoters & enhancers:


Eukaryotic promoters & enhancers

• Promoters located “relatively” close to initiation site

(but can be located within gene, rather than upstream!)

• Enhancers also required for regulated transcription(these control expression in specific cell types, developmental stages, in response to environment)

• RNA polymerase complexes do not specifically recognize promoter sequences directly

• Transcription factors bind first and serve as “landmarks” for recognition by RNA polymerase complexes


Eukaryotic transcription factors

• Transcription factors (TFs) are DNA binding proteins that also interact with RNA polymerase complex to activate or repress transcription

• TFs contain characteristic “DNA binding motifs”

http://www.ncbi.nlm.nih.gov/books/bv.fcgi?rid=genomes.table.7039

• TFs recognize specific short DNA sequence motifs “transcription factor binding sites”• Several databases for these, e.g. TRANSFAC http://www.generegulation.com/cgibin/pub/databases/transfac


Zinc finger-containing transcription factors

• Common in eukaryotic proteins

• Estimated 1% of mammalian genes encode zinc-finger proteins

• In C. elegans, there are 500!

• Can be used as highly specific DNA binding modules


Brown Fig 9.12

• Potentially valuable tools for directed genome modification (esp. in plants) & human gene therapy


New Today: Promoter Prediction

Predicting regulatory regions (focus on promoters) Brief review promoters & enhancers

Predicting promoters: eukaryotes vs prokaryotes

Next week: RNA structure & function


Predicting Promoters

• Overview of strategies What sequence signals can be used?• What other types of information can be used?

• Algorithms • Promoter prediction software

• 3 major types• many, many programs!


Promoter prediction: Eukaryotes vs prokaryotes

Promoter prediction is easier in microbial genomes

Why? Highly conservedSimpler gene structuresMore sequenced genomes!

(for comparative approaches)

Methods? Previously, again mostly HMM-based Now: similarity-based. comparative

methodsbecause so many genomes

available


Predicting promoters: Steps & Strategies

Closely related to gene prediction! • Obtain genomic sequence• Use sequence-similarity based comparison

(BLAST, MSA) to find related genesBut: "regulatory" regions are much less well-conserved than coding regions

• Locate ORFs • Identify TSS (if possible!)• Use promoter prediction programs • Analyze motifs, etc. in sequence (TRANSFAC)


Predicting promoters: Steps & Strategies

Identify TSS --if possible?• One of biggest problems is determining exact TSS!

Not very many full-length cDNAs!• Good starting point? (human & vertebrate genes)

Use FirstEFfound within UCSC Genome Browseror submit to FirstEF web server



Automated promoter prediction strategies

1)Pattern-driven algorithms

2)Sequence-driven algorithms

3)Combined "evidence-based"

BEST RESULTS? Combined, sequential


Promoter Prediction: Pattern-driven algorithms

• Success depends on availability of collections of annotated binding sites (TRANSFAC & PROMO)

• Tend to produce huge numbers of FPs

• Why? • Binding sites (BS) for specific TFs often variable• Binding sites are short (typically 5-15 bp)• Interactions between TFs (& other proteins)

influence affinity & specificity of TF binding • One binding site often recognized by multiple BFs • Biology is complex: promoters often specific to

organism/cell/stage/environmental condition


Promoter Prediction: Pattern-driven algorithms

Solutions to problem of too many FP predictions?

• Take sequence context/biology into account • Eukaryotes: clusters of TFBSs are common• Prokaryotes: knowledge of factors helps

• Probability of "real" binding site increases if annotated transcription start site (TSS) nearby • But: What about enhancers? (no TSS nearby!)

& Only a small fraction of TSSs have been experimentally mapped

• Do the wet lab experiments! • But: Promoter-bashing is tedious


Promoter Prediction: Sequence-driven algorithms

• Assumption: common functionality can be deduced from sequence conservation• Alignments of co-regulated genes should highlight

elements involved in regulationCareful: How determine co-

regulation?• Orthologous genes from difference

species• Genes experimentally determined to be

co-regulated (using microarrays??)• Comparative promoter prediction:

"Phylogenetic footprinting" - more later….


Problems:• Need sets of co-regulated genes• For comparative (phylogenetic) methods

• Must choose appropriate species• Different genomes evolve at different rates• Classical alignment methods have trouble with translocations, inversions in order of functional

elements• If background conservation of entire region is highly

conserved, comparison is useless• Not enough data (Prokaryotes >>> Eukaryotes)

• Biology is complex: many (most?) regulatory elements are not conserved across species!

Promoter Prediction: Sequence-driven algorithms


Examples of promoter prediction/characterization software

Lab: used MATCH, MatInspectorTRANSFACMEME & MASTBLAST, etc.

Others?FIRST EFDragon Promoter Finder (these are links in PPTs)

also see Dragon Genome Explorer (has specialized promoter software for GC-rich DNA, finding CpG islands, etc)JASPAR


TRANSFAC matrix entry: for TATA box

Fields:• Accession & ID •Brief description•TFs associated with this entry•Weight matrix •Number of sites used to build (How many here?)•Other info



Global alignment of human & mouse obese gene promoters (200 bp

upstream from TSS)



Check out optional review & try associated tutorial:

Wasserman WW & Sandelin A (2004) Applied bioinformatics for identification of regulatory elements. Nat Rev Genet 5:276-287


Check this out: http://www.phylofoot.org/NRG_testcases/


Annotated lists of promoter databases & promoter prediction software

• URLs from Mount Chp 9, available onlineTable 9.12 http://www.bioinformaticsonline.org/links/ch_09_t_2.html

• Table in Wasserman & Sandelin Nat Rev Genet article http://proxy.lib.iastate.edu:2103/nrg/journal/v5/n4/full/nrg1315_fs.htm

• URLs for Baxevanis & Ouellette, Chp 5:http://www.wiley.com/legacy/products/subject/life/bioinformatics/ch05.htm#links

More lists:• http://www.softberry.com/berry.phtml?

topic=index&group=programs&subgroup=promoter• http://bioinformatics.ubc.ca/resources/links_directory/?subcategory_id=104• http://www3.oup.co.uk/nar/database/subcat/1/4/


Reading Assignment (for Mon)

Mount Bioinformatics• Chp 8 Prediction of RNA Secondary Structure • pp. 327-355• Ck Errata:

http://www.bioinformaticsonline.org/help/errata2.html

Cates (Online) RNA Secondary Structure Prediction Module• http://cnx.rice.edu/content/m11065/latest/

10/26/05 promoter prediction (really!)

Documents