Exploring the transcriptome

1
Exploring the transcriptome

• From transcripts to genes

2
Transcriptome sequencing

•  Traditional  sequencing
• cDNA clones isolated on the basis of some
  functional property of interest to a group
• EST sequencing
• Large-scale sampling of end sequences of all cDNA
  clones present in a library
•  Full-length  sequencing
• Systematic attemps to obtain high-quality
  sequences of cDNA clones representing all
  transcribed genes

3
Why EST sequencing?

• Systematic sampling of the transcribed portion of
  the genome (transcriptome)
• Provides sequence tags allowing unique
  identification of genes (e.g. for SAGE)
• Provides experimental evidence for the positions
  of exons
• Provides regions coding for potentially new
  proteins
• Provides clones for DNA microarrays

4
ESTs - the basics

• cDNA libraries prepared from various tissues and
  cell lines, using directional cloning
• Gridding of individual clones using robots
• For each clone, sequencing of both ends of insert
  in single pass
• Deposit readable part of sequence in database

5
cDNA libraries

• Most are native, meaning that clone frequency
  reflects mRNA abundance
• Most are primed with oligo(dT), meaning that 3
  ends are heavily represented
• The complexity of libraries is extremely variable
• Normalized libraries are used to enrich for
  rare mRNAs

6
cDNA libraries used

• Large number of libraries represented
• Most libraries managed by the IMAGE consortium
• Human and mouse libraries most abundantly
  represented at IMAGE
• Systematic effort to make libraries from
  cancerous tissue CGAP project (NCI)
• Many tissues still not sampled
• Quality very uneven

7
Clone availability

• In principle, all clones produced by IMAGE are
  publicly available
• Distributors ATCC, Incyte and Invitrogen in US,
  HGMP (UK) and RZPD (Germany) in Europe
• Error rate is high about 30 chance that clone
  doesnt have expected sequence
• Invitrogen sells sets of sequence verified clones

8
EST databases

• EMBL/GenBank have separate sections for EST
  sequences
• ESTs are the most abundant entries in the
  databases (gt60)
• ESTs are now separated by division in the
  databases
• EMBL human, mouse, plant, prokaryote, etc.
• EST sequences are submitted in bulk, but do have
  to meet minimal quality criteria

9
EST entries in EMBL (1)
ID AI242177 standard RNA EST 581 BP. AC
AI242177 SV AI242177.1 DT 05-NOV-1998 (Rel.
57, Created) DT 03-MAR-2000 (Rel. 63, Last
updated, Version 3) DE qh81g08.x1
Soares_fetal_liver_spleen_1NFLS_S1 Homo sapiens
cDNA DE clone IMAGE1851134 3' similar to
gbM10988 TUMOR NECROSIS FACTOR DE PRECURSOR
(HUMAN), mRNA sequence. RN 1 RP 1-581 RA
NCI-CGAP RT National Cancer Institute, Cancer
Genome Anatomy Project (CGAP), Tumor RT Gene
Index http//www.ncbi.nlm.nih.gov/ncicgap RL
Unpublished. DR RZPD IMAGp998P154529
IMAGp998P154529. CC On May 19, 1998 this
sequence version replaced gi2846208. CC
Contact Robert Strausberg, Ph.D. CC Tel (301)
496-1550 CC Email Robert_Strausberg_at_nih.gov CC
This clone is available royalty-free through
LLNL contact the CC IMAGE Consortium
(info_at_image.llnl.gov) for further information. CC
Insert Length 1280 Std Error 0.00 CC Seq
primer -40UP from Gibco CC High quality
sequence stop 463.
10
EST entries in EMBL (2)
FH Key Location/Qualifiers FH FT
source 1..581 FT
/db_xreftaxon9606 FT
/db_xrefESTLIB452 FT
/db_xrefRZPDIMAGp998P154529 FT
/noteOrgan Liver and Spleen Vector pT7T3D
(Pharmacia) FT with a modified
polylinker Site_1 Pac I Site_2 Eco RI FT
This is a subtracted version of the
original Soares fetal FT liver
spleen 1NFLS library. 1st strand cDNA was
primed FT with a Pac I -
oligo(dT) primer 5' FT
AACTGGAAGAATTAATTAAAGATCTTTTTTTTTTTTTTTTTTT
3', FT double-stranded cDNA
was ligated to Eco RI adaptors FT
(Pharmacia), digested with Pac I and cloned
into the Pac I FT and Eco RI
sites of the modified pT7T3 vector. Library FT
went through one round of
normalization. Library FT
constructed by Bento Soares and M.Fatima
Bonaldo. FT /sexmale FT
/organismHomo sapiens FT
/cloneIMAGE1851134 FT
/clone_libSoares_fetal_liver_spleen_1NFLS_S1 FT
/dev_stage20 week-post
conception fetus FT
/lab_hostDH10B (ampicillin resistant)
11
EST sequence quality

• Single, unverified runs, so quality is on average
  low (and worsening towards 3)
• Current submission rules require Phred score gt
  20 (lt1 error) documented
• Trivial contaminants are common (vector, rRNA,
  mitRNA, other species )
• Frameshift errors are very common
• For many reads, trace files can be retrieved

12
Private EST databases

• Producing and selling access to EST data has
  proven to be a lucrative business
• Incyte has created the LifeSeq databases, to
  which access can be bought
• Human Genome Sciences has preferred to exploit
  the data itself, and to get patents on promising
  genes found in its databases

13
Why search EST databases?

• Alternative to library screening a short tag can
  lead to a cDNA clone
• Alternative to cDNA sequencing sequences of
  multiple ESTs can reconstitute a full-length cDNA
• Gene discovery EST sequences often contain new
  members of known families

14
BLAST searching EST databases

• Many sites where EST databases can be searched
  with BLAST, including our own
• Searching raw EST data is becoming less
  attractive, as for many species ESTs have been
  incorporated into transcript assemblies
• There are at present no  consensus  transcript
  assemblies

15
Problems with raw EST databases

• The databases are highly redundant e.g. 5x106
  human sequences for 4x104 genes
• The databases are skewed for sequences near 3
  end of mRNAs
• The error rates are high in individual ESTs
• For most ESTs, there is no indication as to the
  gene from which is was derived

16
The ORESTES project

• Goal to obtain EST sequences from the
  under-represented, often coding, central portions
  of mRNAs
• Methodology use low-stringency semi-random
  priming followed by PCR, producing low complexity
  libraries
• Results over 1000000 ESTs produced, of which
  half produce novel information

17
How to organize EST collections?

• Clustering associate individual EST sequences
  with unique transcripts or genes
• Assembling derive consensus sequences from
  overlapping ESTs belonging to the same cluster
• Mapping associate ESTs (or EST contigs) with
  exons in genomic sequences
• Interpreting find and correct coding regions

18
The Unigene database

• Unigene is an ongoing effort at NCBI to cluster
  EST sequences with traditional gene sequences
• For each cluster, there is a lot of additional
  information included
• Unigene is regularly rebuilt. Therefore, cluster
  identifiers are not stable gene indices

19
The TIGR Gene Indices

• The Gene Indices are based on contig assemblies
  rather than true clustering strategies
• There are more Gene Indices than Unigene clusters
• TIGR maintains EGAD, a database of known human
  transcripts
• EGAD entries are included in the Human Gene
  Indices

20
Other EST cluster and assembly databases

• SANBI in South Africa produces the STACK
  collection of human EST contigs
• MIPS in Munich and the SIB produce
  BLAST-searchable contigs from Unigene
• TIGEM in Italy has a nice collection of EST
  search and assembly tools, local remote
• The CBIL at the U. of Pennsylvania has assembled
  the DOTS database, part of the AllGenes project

21
Strategy for gene discovery

• Cluster EST sequences to identify candidate
  transcripts
• Assemble to increase length and reduce redundancy
• Find coding regions and reading frame, and
  correct frameshift error
• Use deduced protein sequences as searchable
  database (TrEST)

22
Interpreting ESTs in the context of the genome

• The genome sequence provides a reliable scaffold
  for clustering ESTs
• ESTs clustered on the genome can be assembled by
  following exon/intron boundaries
• The combination of genome, full-length insert and
  EST data will provide the reference dataset for
  gene and transcript annotation

23
Using EST data for gene expression analysis

• Basic idea compare relative abundance of
  specific cDNAs in libraries of different origins
• Best tools available Digital Gene Expression
  Displayer (DGED) and xProfiler at the CGAP Web
  site
• Alternative browse through genes unique to
  different libraries

24
Full-length cDNA sequencing

• Principle make high-quality cDNA libraries,
  often optimized for full-length inserts
• Do first-pass sequencing of 3 and 5 ends, and
  select good clones one per transcript,
  contains initiation codon and poly(A)
• For selected clones, fully sequence the inserts

25
Full-length sequencing projects

• MGC (at NCI) goal of 20000 sequences a year
  starting in 2001, human and mouse actually
  produced 15000 human and 11000 mouse
• RIKEN Genome Science Lab, Tsukuba produced and
  annotated 60000 full-length mouse sequences
  (FANTOM2)
• Kazusa Institute HUGE project set of over 2000
  KIAA human cDNA sequences (large proteins)
• NEDO human cDNA project over 20000 FLJ
  sequences available in early 2003
• German cDNA consortium (DKFZ) produced over 3200
  sequences, current state of project unclear

26
The future of ESTs

• In human and mouse, most will come as byproducts
  of full-length projects,
• There are good arguments for trying to reach
  saturation on selected tissues
• ESTs are still the tool of choice for rapid
  exploration of the transcriptomes of various
  species, especially with large genomes
• ESTs could form a very solid basis for
  evolutionary studies

27
The need for a gene index

• All high-throughput biology methods require a
  unique and reliable way to describe the genes
  they are analysing
• This index should be stable, unique, extensible,
  and independent of a system of nomenclature
• The index should document all transcript
  sequences belonging to the corresponding gene

28
Some commonly used indices

• EMBL/GenBank/DDBJ accession numbers
• Unique and universally accepted BUT
• Highly redundant (many entries per gene)
• Unigene cluster identifiers (NCBI)
• Widely used and non-redundant BUT
• Rely on clustering procedure (unreliable) AND
• Unstable clusters change with each build
• RefSeq accession numbers (NCBI)
• Stable and non-redundant BUT
• Still very far from comprehensive AND
• Many RefSeq sequences are incomplete AND
• Splice variants are not systematically documented

29
Emerging gene indices

• HUGO Gene Nomenclature Committee
• Assigns unique symbols to human genes with a
  known or putative function
• About 15000 symbols assigned so far
• No attempt to link to sequence data or variants
• Ensembl
• Uses  conservative  pipeline to predict genes
  and transcripts
• 25000 predicted genes and 37000 transcripts so
  far
• Provides unique identifiers for genes and
  transcripts
• Linked to HGNC names when possible

30
Gene documentation resources

• Repositories collecting info about loci from
  multiple sources
• LocusLink (NCBI) links mostly to NCBI internal
  information sources
• GeneCards (Weizmann Inst.) uses Web robots to
  collect information, probably most complete
  resource available today
• GeneLynx (Karolinska Inst.) similar to
  GeneCards, but also collects EST information

31
 tromer  goals

• To combine the draft genome and transcriptome
  into a database documenting all human transcripts
• To design a rational representation and indexing
  system for human genes
• To reconstitute transcripts and their variants
  using genomic DNA as template

32
Methods

• Create a comprehensive and non-redundant set of
  mRNA 3 ends, documented by polyadenylation sites
  mapped to the genome
• Create a comprehensive mapping of EST sequences
  to exons on the genome
• Use this to create a database of exons and
  introns, defined by their coordinates on the
  genome
• Combine these two data sets to reconstitute
  transcripts

33
Poly(A) site documentation

• Re-extract EST sequences from trace files (when
  available)
• Extract 50 nt adjacent to poly(A)
• Make this collection non-redundant
• Map all tags to the draft genome
• Validate the quality of the tags

34
Generation of the 3 tags
35
Example of a  bad  tag
ID 3P016851 SQ 1 0 AGCACTGATGTTGAATTTCACTTTA
AGTTGTCACATGGTCTGAGTTGTACaaaaagaaagaaaaattctggccgg
gtgtggtggctcatgcctgtaatcc PA Score-315
matchnegative QQ A13 C0 G2 T0
testbad CM nz76g10.s1AA764893 -, 18 As in
polyA CO qt57h10.x1AI355333 GN AP000848_32
pos 981 CHR 11 GN AP001320_19 pos
2151 CHR 11 GN AC011875_21 pos 1166 CHR
NA //
Example of a  good  tag
ID 3P016857 SQ 1 0 AGCACTGTCTTATCACATCGCCAAT
TAGTTGTAATAAACGTTCAACGTACaaacactggagtgtgtttttatctc
tttccaaaagtttgtcaaactatgc PA Score 117
matchstrong QQ A5 C2 G5 T3 testok CM
nk75b07.s1AA551090 -, 17 As in polyA CO
ob29c06.s1AA741544 qm22a02.x1AI278652
wj07f01.x1AI824173 wm06d04.x1AI887507 CO
xf19g12.x1AW131962 yw01c11.s1H96551
yy23c08.s1N35379 zl39e04.s1AA149587 CO
zq01c05.s1AA193118 zv54a07.s1AA437212 GN
AC004106_1 - pos 34804 CHR X SQ 2 4
CTGTCTTATCACATCGCCAATTAGTTGTAATAAACGTTCAACGTACAAAC
actggagtgtgtttttatctctttccaaaagtttgtcaaactatgcagag
PA Score 134 matchstrong QQ A2 C1 G5
T7 testok CM aa28a05.r1AA480914 -, 16 As
in polyA CO qe01d07.x1AI143977
qu50g08.x1AI285755 yh71a05.s1R32806
yv79c01.r1H82355 CO zn91a12.s1AA127331
zn92a12.s1AA127424 GN AC004106_1 - pos
34800 CHR X //
36
Exon and intron mapping

• Document all matches between genome contigs and
  RNAs (pairwise matching) using BLAST (gt5x106
  matches)
• Create pairwise alignments between RNAs and
  genome contigs using sim4, document exon and
  intron boundaries
• Import data into ACEDB for exploration and
  visualization

37
gtchrAC018734_9AC018734.3 Chromosome 21 Clone
RP11-351D2 SC WashU.Homo sapiens
LEN30562 gtestAW468733AW468733.1-
LEN413 16806-16850 (1-45) 97 -gt 18234-18383
(46-195) 100 -gt 19199-19266 (196-263) 100
-gt 19463-19612 (264-413) 100 0 .
. . . .
16806 CCCGTCGGGAGACACGTTTTGTCAGTTGCTTCCTCTGGAGTCCT
TGTG..
gtgtgt.. 1 CCNGTCGGGAGACACGTTTTGTCAGT
TGCTTCCTCTGGAGTCCTT 50 . .
. . . 16856
.AAGTTAAAAGACTACTTTTTAGAGCAGTTTTCATTTTACCGATACATCG
.gtgtgt
46 TTAAAAGACTACTTTTTAGAGCAGTTTT
CATTTTACCGATACATCG 100 . .
. . . 18280
ATCAGAGAGTACGGAGCTTCCCACATTGCCTCCCGCACCCCATAGCAGGC

92 ATCAGAGAGTACGGAGCTTCCCACATTGCCTC
CCGCACCCCATAGCAGGC 150 . .
. . . 18330
AGAGGCACCAGCCAGGCGCCATCCTTTACAAGCTGTTTCAGCAAAATGTC

142 AGAGGCACCAGCCAGGCGCCATCCTTTACAAG
CTGTTTCAGCAAAATGTC 200 . .
. . . 18380
CCTGGTG...CAGGGTCTTAATGATTAAGTCCAGGCACTCCAAGGACGAT
gtgtgt...gtgtgt
192 CCTG GGTCTTAATGATTAAGTCC
AGGCACTCCAAGGACGAT 250 . .
. . . 19236
TAAATATTTAAACAGCAGCTTCCCTTTGAAGGTC...CAGGGAAGGAAAT
gtgtgt...gtgtgt
233 TAAATATTTAAACAGCAGCTTCCCTTTGAAG
GGAAGGAAAT 300 . .
. . . 19473
GACCGCAGAAGGGAGTGCACCGTGGGGGCAGGAGGTGCTGCGGCCAATGG

274 GACCGCAGAAGGGAGTGCACCGTGGGGGCAGG
AGGTGCTGCGGCCAATGG 350 . .
. . . 19523
GCAGTGCTGGGAGATGAAACCAACAGCGGCGATTGCCCCAGAAAGAACTT

324 GCAGTGCTGGGAGATGAAACCAACAGCGGCGA
TTGCCCCAGAAAGAACTT 400 . .
. . 19573 GGCGCTGCTAAGGCAATGGCGGC
CCGAAAGAGGCCCCTGA
374 GGCGCTGCTAAGGCAATGGCGGCCC
GAAAGAGGCCCCTGA
Output of sim4 program
38
NCAM2
Hs.177691
Hs.135892
DKFZp761I1311
Hs.76118
39
COL18A1
Hs.78409
3 UTR overlap
Hs. 84190
SLC19A1
40
Alternative polyadenylation

• Affects at least half of all genes
• Estimate based on manual examination of 3.5 MB
  contig from chr 21
• Exists at both micro (lt100 bp) and macro (gt100
  bp) levels
• Often extends over distances of several kilobases
• Impacts on microarray probe design as well as
  SAGE data interpretation

41
Reconstituting transcripts (1)

• Data cleanup
• Decide which transcript to genome matches are
  reliable (gene families, pseudogenes, genome
  repeats, etc)
• Within alignments, decide which parts are bona
  fide exons
• Exon generation
• Determine directionality, based on splice
  junctions and RNA annotations
• Determine exon boundaries and completeness

42
Reconstituting transcripts (2)

• Merge multiple occurrences of same exon,
  respecting boundary rules
• E.g. overlapping exons with different splicing
  patterns are not merged
• Merge multiple occurrences of same splice
  junctions
• Resolve differences due to noise in data from
  real alternative splice junctions

43
A
3tags
9,995
Alu
3,592
3,674
5,734
5,779
6,122
6,152
6,267
6,355
6,577
6,749
8,162
8,505
1
2
3
4
5
6
7
C
B
1
2
3
4
5
6
E
A
E
7
44
Genes as graphs
B0 5600 5/1 gt D0 5/1 D0 6769
5/1 B0 5/1 gt -gt A0 5/1 A0 8755
5/1 D0 5/1 -gt gt D1 5/1 D1 9013
5/1 A0 5/1 gt -gt A1 5/1 B1 10515
2/1 gt D2 2/1 A1 10929 5/1 D1 5/1
-gt gt D2 5/1 D2 11092 7/2 B1 2/1 gt
A1 5/1 gt -gt A2 7/2 A2 11673 10/1
D2 10/1 -gt gt E0 10/1 E0 12486 10/1
3P008703 A2 10/1 gt

• Nodes are connected by edges
• Edges are exons or introns
• Nodes are- B initiation- D splice donor-
  A splice acceptor- E polyadenylation site

B0
D0
A0
D1
B1
A1
D2
A2
E0
45
Reconstituting transcripts (3)

• Generate graphs for each gene in a genomic
  segment
• A gene is a fully connected graph
• Graphs contain nodes and edges linking these
  nodes (exons)
• Generate the most likely transcripts for each
  gene
• A transcript is a path through the graph
• The decision on which paths to keep or reject is
  based on the frequencies with which edges are
  documented

46
Towards a gene index?

• The graph nodes are sequence based and can easily
  be made non-redundant
• The connectivity between nodes can be completed
  as more data become available
• The gene index derived from graphs includes a
  full representation of the locus, including all
  splice variants and all of the primary transcript
  data (ESTs, full-length)

47
What are the uses of the graphs?

• Solid experimental basis for designing microarray
  probes (3 ends)
• Mapping of SAGE and MPSS tags
• Exploration of the diversity of protein coding
  potential (alternative splicing)
• Gene discovery based on expression patterns,
  either for whole locus or exon by exon
• Exploration of the complexity of genes