---

a generic tool for sequence alignment

---

A beginner's guide to using exonerate

• This page gives some examples of using exonerate to perform various types of pairwise comparison.
• After this, you may wish to look at the advanced guide.
• Remember to use -h to get a short summary of available options, or --help for a longer summary.
• Further information can be found on the exonerate man page.
• If you have questions, or more examples to add to this tutorial, please email me: guy@ebi.ac.uk

The examples:

1. Simple ungapped alignment
2. Gapped alignment against a database
3. Aligning a cDNA to genomic sequence
4. Aligning a protein to genomic sequence
5. 6-frame translated alignment"
6. Genome to genome alignment
7. Exhaustive Smith-Waterman-Gotoh alignment
8. Alignment including non-equivalenced regions

Simple ungapped alignment

• Ungapped alignment is the default for exonerate so you only need to supply the query and target sequences:

  exonerate query.fasta target.fasta
  

• Its that simple

Gapped alignment against a database

• To generate gapped alignments, we have to specify an alignment model. The model used here is an affine local alignment model, which is equivalent to the classic Smith-Waterman-Gotoh type of alignment.

  exonerate --model affine:local query.fasta target.fasta
  

Aligning a cDNA to a genomic sequence

• To align a cDNA to genomic sequence, use the est2genome model.

  exonerate --model est2genome query.fasta target.fasta
  

• This will allow introns to be placed in the alignment, allowing exons on the genomic sequence to align to the query.
• There are many other models which can be used. See the man page for a complete list.

Aligning a protein to genomic sequence

• Similarly, it is possible to align a protein sequence to the genome, (similar to GeneWise, but with heuristics).

  exonerate --model protein2genome query.fasta target.fasta
  

• This model will allow introns in the alignment, but also allow frameshifts, and exon phase changes when a codon is split by an intron.

6-frame translated alignment

• The coding2coding model aligns two DNA sequences by translated of both the query and target in all reading frames, while allowing for gaps and frameshifts.

  exonerate --model coding2coding query.fasta target.fasta
  

Genome to genome alignment

• When aligning very large sequences, (chromosome-sized), using the --bigseq option allows you to generate genome alignments quickly and .
• Then doing this it is also a good idea to raise the hsp threshold to reduce the number of HSPs produced.

  exonerate -m a:l --bigseq yes --dnahspthreshold 120 
            chr21.fasta chr22.fasta
  

• This can be done with any alignment model.

Exhaustive Smith-Waterman-Gotoh alignment

• By default, exonerate will use heuristics to generate alignments, however for the maximum sensitivity, exhaustive alignment should be used.

  exonerate -m a:l --exhaustive yes query.fasta target.fasta
  

• Note this requires quadratic time, so will be slow for very long sequences (although it will only require linear memory).

Alignment including non-equivalenced regions

• Another model provided by exonerate is the NER model, which generates alignments where colinear blocks are joined by non-equivalenced (unaligned) regions.

  exonerate -m ner query.fasta target.fasta