ASTD

1
ASTD

• Alternative Splicing and Transcript Diversity
  database

2
What/who are we?

• Firstly AltExtron
• Secondly ASD - Alternative splicing database, and
  the AltSplice pipeline
• database of alternative splice events and the
  resultant isoform splice patterns of genes from
  human, and other model species.
• Thirdly, for grant purposes, ATD - Alternative
  transcript diversity database, and the AltTrans
  pipeline
• formation of transcript isoforms on a genome-wide
  scale by creating a value-added database of
  full-length alternate transcripts from human and
  other model species.
• We also host the AEdb database manual
  annotations
• the two, ASD and ATD, blended into 1 pipeline, so
  now we are
• ASTD
• Alternative splicing and transcript diversity
  database
• www.ebi.ac.uk/astd

3
Pipeline in a nutshell
Poly(A) Pipeline
1. Ensembl gene slices EMBL
EST/mRNA/HTC/HInv download
TSS Pipeline
Peptide Pipeline
2. Immunoglobulin filtering (Blast)
9. Data generation
SNP Pipeline
3. Redundant gene filtering (Blat)
8. Events prediction
Conservation Pipeline
4. Genes vs EST/mRNA Alignment (Blast)
7. Splice patterns delineation
6. Intron/exon delineation
5. HSP Collection
4
Limitations of the pipeline

• Pipeline defines consensus splice sites
• True biology is removed
• dicistronic transcripts
• Nested genes
• Single exon genes
• Small exons
• Large introns
• Manual annotation would resolve these issues ..

5
Improvements

• New web interfaces user friendly
• New database schema that is normalised,
  extendable and maintainable
• Pipeline improvements some steps now automated,
  bugs corrected, some improvements and blat
  replaces blast for filtering redundant genes
• Database allows external features to be included
  (Ensembl and VEGA annotations) to compare to our
  transcripts
• Schema allows export of data in standard format
  GTF2 and GFF3, EMBL flat file format, fasta
  format, and excel spreadsheet
• Transcripts for complete genome, not restricted
  to those with alternative splice events
• Introduction of unique identifiers
• Addition of datasets as input to pipeline HTC
  and HInv
• Extension of 5 and 3 UTR to capture more TSS
  and poly(A)
• Annotation of TSS (Align 5 capped mRNAs from
  human and mouse to transcript ) and poly(A) to
  generate full length transcripts

6
www.ebi.ac.uk/astd - Query tools

• Three query tools are available to retrieve
  entries
• Simple text search on the main page
• Genome browsing
• Advanced search

7
Gene information
8
Genomic region information 1
9
Genomic region information 2
10
Transcript information
11
evidence for transcript 1
12
evidence for transcript 2
13
Expression information
14
Splice event 1
15
Splice event 2
16
Peptide information
17
Statistics

• Human
• Number of genes with an ASTD transcript
  16715
• Number of genes with an ASTD transcription_start
  _site 4936
• Number of genes with an ASTD polyA_site
  15376
• Number of genes with an ASTD splicing event
  11316
• Number of genes with multiple ASTD transcripts
  14101
• Proportion of genes undergoing alternative
  splicing 68
• Proportion of genes undergoing alternative
  polyadenylation 92
• Proportion of genes undergoing alternative
  transcription_start_sites 30
• Mouse
• Number of genes with an ASTD transcript
  16491
• Number of genes with an ASTD transcription_start
  _site 948
• Number of genes with an ASTD polyA_site
  13556
• Number of genes with an ASTD splicing event
  9474
• Number of genes with multiple ASTD transcripts
  13028
• Proportion of genes undergoing alternative
  splicing 57
• Proportion of genes undergoing alternative
  polyadenylation 82
• Proportion of genes undergoing alternative
  transcription_start_sites 6

18
Graph of human growth
19
Controlled vocabularies/ontologies

• GO
• SOFA
• eVOC
• Splice event ontology
• MeSH terms

20
Future 1

• Addition of new species
• Experimental validation of transcript structure
  and alternative poly(A)s
• Use EMBL CDS as another source of alignments to
  the genome
• More frequent releases every 3 months
• Addition of regulatory motifs ESS, ESE, ISS and
  ISE
• microRNA target sites from the EURASNET NoE
  (University Basel)

21
Future 2

• Introduction of unique identifiers means
• Addition as xref in EMBL so transcripts in the
  INSDC can be grouped into one gene
• Addition into UniParc so translations can be
  linked to UniProt IsoIds and again grouped as
  being variants of one gene
• UniParc translations also undergo full InterPro
  scan, TM and SignalP predictions so data is
  precomputed and not done on the fly

22
Future 3

• The EBI sequence database group and Ensembl have
  merged making the Hinxton Sequencing Forum (HSF)
• Outcome is that ASTD will be vehicle to augment
  the Ensembl transcript views
• Full length transcripts with TSS, splice events
  and polyA
• Definition of the major transcript set using
  annotation of features to transcripts, eg
  expression state, exon array, splice junction
  array evidence etc
• VEGA/Havana annotations also included
• Time scale - within 2 years

23
Acknowledgements

• The ASTD Team
• Gautier Koscielny
• Vincent Le Texier
• Eleanor Whitfield
• Chellapa Gopalakrishnan
• Vasudev Kumanduri
• Sequence Database Group and External Services
• ASD consortium (Stefan Stamm for AEdb)
• ATD consortium (Daniel Gautheret for AltPAS)
• EURASNET consortium
• The ASTD project at EBI is supported by a grant
  from the EC Eurasnet Network of Excellence
  (LSHG-CT-2005-518238).