Genome Annotation

Introduction to Genome Annotation
AGCGTGGTAGCGCGAGTTTGCGAGCTAGCTAGGCTCCGGATGCGA
CCAGCTTTGATAGATGAATATAGTGTGCGCGACTAGCTGTGTGTT
GAATATATAGTGTGTCTCTCGATATGTAGTCTGGATCTAGTGTTG
GTGTAGATGGAGATCGCGTAGCGTGGTAGCGCGAGTTTGCGAGCT
AGCTAGGCTCCGGATGCGACCAGCTTTGATAGATGAATATAGTGT
GCGCGACTAGCTGTGTGTTGAATATATAGTGTGTCTCTCGATATG
AGTCTGGATCTAGTGTTGGTGTAGATGGAGATCGCGTGCTTGAGT
TCGTTCGTTTTTTTATGCTGATGATATAAATATATAGTGTTGGTG
GGGGGTACTCTACTCTCTCTAGAGAGAGCCTCTCAAAAAAAAAGC
CGGGGATCGGGTTCGAAGAAGTGAGATGTACGCGCTAGCTAGTAT
ATCTCTTTCTCTGTCGTGCTGCTTGAGATCGTTCGTTTTTTTATG
GATGATATAAATATATAGTGTTGGTGGGGGGTACTCTACTCTCTC
AGAGAGAGCCTCTCAAAAAAAAAGCTCGGGGATCGGGTTCGAAGA
AGTGAGATGTACGCGCTAGXTAGTATATCTCTTTCTCTGTCGTGC
What is Annotation?
• dictionary definition of “to annotate”:
– “to make or furnish critical or explanatory notes or comment”
• some of what this includes for genomics
– gene product names
– functional characteristics of gene products
– physical characteristics of gene/protein/genome
– overall metabolic profile of the organism
• elements of the annotation process
– gene finding
– homology searches
– functional assignment
– ORF management
– data availability
• manual vs. automatic
– automatic = computer makes the decisions
• good on easy ones
• bad on hard ones
– manual = human makes the decisions
• highest quality
**Due to the VOLUMES of genome data today, most genome projects are
annotated primarily using automated methods with limited manual annotation
Annotation pipeline
Generation of Open Reading Frames
Homology Searches
Putative ID
Frameshift Detection
Ambiguity Report
Role Assignment
Metabolic Pathways
Gene Families
DNA Motifs
Regulatory Elements
Repetitive Sequences
Comparative Genomics
Genome Structure
Prokaryote
Intergenic Monocistronic Polycistronic

Region Gene Intergenic Genes in an
Region Operon
Eukaryote
Gene Intergenic Gene Intergenic Gene

Region Region
Prokaryotic Gene Structure and Transcript Processing
http://pps00.cryst.bbk.ac.uk/course/section6/henryb/genestrp.htm
Eukaryotic Gene Structure and Transcript Processing
Structural Annotation: Finding the
Genes in Genomic DNA
Two main types of data used in defining gene
structure:
Prediction based: algorithms designed to find

genes/gene structures based on nucleotide
sequence and composition
Sequence similarity (DNA and protein): alignment

to mRNA sequences (ESTs) and proteins from the
same species or related species; identification of
domains and motifs
Finding Genes (ORFs)
Gene finders a programs that can identify genes computationally
Running a Gene-finder
is a two-part process
1) Train Gene finder for the organism you

have sequenced.
2) Run the trained Gene finder on the

completed sequence.
Candidate Genes
6-frame ORF map
Stop codons (TAA, TAG, TGA) (long hash marks)
Start codons (ATG, GTG, TTG) (short hash marks)
+3
+2
+1
-1
-2
-3
Minimum ORF Length
ORFs over minimum length highlighted

Annotating ORFs
Possible translations represented by arrows, moving from start to stop, the dotted
line represents an ORF with no start site.
Glimmer chooses the set of likely genes.
ORF00001 ORF00002 ORF00003 ORF00004

Eukaryotic Gene Finding
Identifying the protein coding region of genes
AAAGCATGCATTTAACGAGTGCATCAGGACTCCATACGTAATGCCG
Gene finder (many

different programs)
AAAGC ATG CAT TTA ACG A GT GCATC AG GA CTC CAT ACG TAA TGCCG
*This is a eukaryotic gene as evidenced by the intron

Signals Within DNA
• Splice sites to identify intron/exon junctions
• Transcription start and stop codons
• Promoter regions
• PolyA signals
Experimental Evidence
DNA sequence evidence: Transcript

sequence (EST, full length cDNA, other
expression types); more restrictive in
evolutionary terms
Protein Evidence: alignment to protein that

suggests structural similarity at the amino
acid level; can be more distant
evolutionarily
Experimental Evidence
Transcript evidence:
-Demonstrates gene is transcribed
-Delineates exon boundaries
-Defines splice sites and alternative transcripts
-If EST based, indicates expression patterns

Functional Assignments
Name
Descriptive common name for the protein, with as much
specificity as the evidence supports; gene symbol.
Role
Describe what the protein is doing in the cell and why.
Associated information:
Supporting evidence:Domain and motifs
EC number if protein is an enzyme.
Paralogous family membership.
Evidence for Gene Function
• PROSITE Motifs
– collection of protein motifs associated with active
sites, binding sites, etc.
– help in classifying genes into functional families
when HMMs for that family have not been built
• InterPro
– Brings together HMMs (both TIGR and Pfam)
Prosite motifs and other forms of motif/domain
clustering
– Results in motif “signatures” for families or functions
– GO terms have been assigned to many of these
Sequence Alignments
Compare sequence
against other databases
Gene function evidence
Functional Annotation: Gene Product Names
Gene Name Assignment: Based on similarity to known proteins in nraa
database
Categories:
Known or Putative: Identical or strong similarity to documented gene(s) in

Genbank or has high similarity to a Pfam domain; e.g. kinase, Rubisco
Expressed Protein: Only match is to an EST with an unknown function; thus

have confirmation that the gene is expressed but still do not know what the
gene does
Hypothetical Protein: Predicted solely by gene prediction programs and

matches another hypothetical or expressed protein
Hypothetical Protein: Predicted solely by gene prediction programs; no

database match
Annotation example
• A good example of this is seen with transporters, what you’ll
see:
– Multiple hits to a specific type of transporter
– -The substrate identified for the proteins your protein matches are
not all the same, but fall into a group, for example they are all
sugars.
• Give the protein a name with specific function but a more

general substrate specificity:
– “sugar ABC transporter, permease protein”
• Sometimes it will not be possible to identify particular substrate

group, in that case:
– “ABC transporter, permease protein”
Automated Annotation
is Not a Solved Problem
What you are getting is output from a series

of prediction tools or alignment programs
• Manual curation is often used to assess various

types of evidence and improve upon automated
gene calls and alignment output
• Ultimately, experimental verification is the only

way to be sure that a gene structure is correct
Structural Annotation: Graphic Viewer Annotation Station
Sequence Database Hits

Top: Protein matches Not shown graphically: gene name, nucleotide and protein sequence, MW, pI,
Bottom: EST matches organellar targeting sequence, membrane spanning regions, other domains.
Gene
Predictions
Annotated Gene
Top: editing panel Splice site predictions:
Bottom: final curation red: acceptor sites
blue: donor sites
Screenshot of a component within Neomorphic’s annotation station: www.neomorphic.com

Features Typically Resolved During Manual Annotation
• incorrect exon boundaries

• merged, split, missing genes
• missing untranslated regions (UTRs)
• missing alternative splicing isoform
annotations
• degenerate transposons annotated as protein-
coding genes
Increasing Complexity of Genome Annotation
# Genes bp
Mycoplasma pulmonis 780 964,000
Escherichia coli K-12 4,300 4,641,000
Saccharomyces cerevisiae 6,300 12,100,000

Plasmodium falciparum 5,400 22,850,000
Caenorhabditis elegans 19,000 97,000,000
Drosophila melanogaster 16,000 120,000,000

Arabidopsis thaliana 25,000 115,400,000
Fugu rubripes 35,000 365,000,000

Homo sapiens 30,000 2,910,000,000
Decrease in gene density and the presence of more, larger introns

Caveats of Genome Annotation
-Greatly impacted by the quality of the sequence; the impact of draft sequencing
on whole genome annotation has yet to be seen by Joe/Jane Scientist. There
will be disappointment when the research communities realize that they don’t
have the “gold” standard of sequence as present in Arabidopsis and rice.
-Annotation is challenging, highly UNDER-estimated in difficulty, highly UNDER-

valued until a community goes to use its genome sequence
-Annotation can be done to high accuracy on a single gene level by single

investigators with expertise in gene families. The challenge is how to extrapolate
this to the whole genome
-Blends of automated, semi-automated, and manual annotation is perhaps the

best way to approach genomes in which there are not large communities
-Iterative, never perfect, can always be improved with new evidence and
improved algorithms

Genome Annotation

Uploaded by

Copyright:

Available Formats

Genome Annotation

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Genome Annotation

Uploaded by

Copyright:

Available Formats

Introduction to Genome Annotation

Generation of Open Reading Frames

Intergenic Monocistronic Polycistronic

Gene Intergenic Gene Intergenic Gene

Prediction based: algorithms designed to find

Sequence similarity (DNA and protein): alignment

Gene finders a programs that can identify genes computationally

1) Train Gene finder for the organism you

2) Run the trained Gene finder on the

Minimum ORF Length

ORFs over minimum length highlighted

Glimmer chooses the set of likely genes.

ORF00001 ORF00002 ORF00003 ORF00004

Identifying the protein coding region of genes

Gene finder (many

*This is a eukaryotic gene as evidenced by the intron

DNA sequence evidence: Transcript

Protein Evidence: alignment to protein that

-Demonstrates gene is transcribed

-Delineates exon boundaries

-Defines splice sites and alternative transcripts

-If EST based, indicates expression patterns

Known or Putative: Identical or strong similarity to documented gene(s) in

Expressed Protein: Only match is to an EST with an unknown function; thus

Hypothetical Protein: Predicted solely by gene prediction programs and

Hypothetical Protein: Predicted solely by gene prediction programs; no

• Give the protein a name with specific function but a more

• Sometimes it will not be possible to identify particular substrate

What you are getting is output from a series

• Manual curation is often used to assess various

• Ultimately, experimental verification is the only

Sequence Database Hits

Screenshot of a component within Neomorphic’s annotation station: www.neomorphic.com

• incorrect exon boundaries

Escherichia coli K-12 4,300 4,641,000

Saccharomyces cerevisiae 6,300 12,100,000

Caenorhabditis elegans 19,000 97,000,000

Drosophila melanogaster 16,000 120,000,000

Fugu rubripes 35,000 365,000,000

Decrease in gene density and the presence of more, larger introns

-Annotation is challenging, highly UNDER-estimated in difficulty, highly UNDER-

-Annotation can be done to high accuracy on a single gene level by single

-Blends of automated, semi-automated, and manual annotation is perhaps the

You might also like