Bioperl course

by Catherine Letondal and Katja Schuerer

Bioperl course [http://www.pasteur.fr/recherche/unites/sis/formation/bioper

by Catherine Letondal and Katja Schuerer Introduction to Bioperl (bio.perl.org [http://bio.perl.org/]). This course introduces to the bioperl modules with examples and exercises. The course content has been upgraded to bioperl 1.0 (differences with bioperl version 0.7 are displayed in yellow color in the diagrams). Apart from this upgrade, the presentation has been reorganized by topics, rather than by daily schedule, without any major changes in the content. A PDF [support.pdf] version is now available.

Table of Contents
1. General introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.1. Bioperl documentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.2. General bioperl classes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 2. Sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2.1. The Bio::SeqIO class . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2.1.1. Format Converter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2.2. Sequence classes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 2.2.1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 2.2.2. Building mechanisms summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 2.2.3. A deeper insight into the Bio::Seq class . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 2.2.4. Bioperl Sequence classes structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 2.3. Features and Location classes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 2.3.1. Feature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 2.3.2. Code reading: extracting CDS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 2.3.3. Tag system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 2.3.4. Location . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 2.3.5. Graphical view of features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 2.4. Sequence analysis tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 3. Alignments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 3.1. AlignIO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 3.2. SimpleAlign . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 3.3. Code reading: protal2dna. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 4. Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 4.1. Blast . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 4.1.1. Running Blast . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 4.1.2. Parsing Blast . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 4.1.3. Bio::Tools::BPlite family parsers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 4.1.4. PSI-BLAST (Position Specic Iterative Blast) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 4.1.5. bl2seq: Blast 2 sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 4.1.6. Blast Internal classes structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 4.2. Genscan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 5. Databases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 5.1. Database classes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 5.2. Accessing a local database with golden . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 6. Perl Reminders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 6.1. UML . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 6.2. Perl reminders to use bioperl modules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 6.2.1. References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 6.2.2. Filehandles and streams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 6.2.3. Exceptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 6.2.4. Getopt::Std . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 6.2.5. Classes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 6.2.6. BEGIN block . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56

6.3. Perl reminders for a further advanced understanding of bioperl modules . . . . . . . . . . . . . . . . . . 6.3.1. Modules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3.2. Compiler instructions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3.3. Tie . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A.1. Sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A.2. Alignments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A.3. Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A.4. Databases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

56 56 56 57 59 59 65 66 78

List of Figures
2.1. Bio::Seq class structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 2.2. Relation of a SwissProt entry and the corresponding Bio::Seq object components . . . . . . . . . . . . . . . . 6 2.3. Bio::Annotation package structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 2.4. Bioperl Sequence classes structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 2.5. Features Classes structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 2.6. Correspondance between an EMBL entry and bioperl tags . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 2.7. Location Classes Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 2.8. Graphical view of some features of the SwissProt entry BACR_HALHA . . . . . . . . . . . . . . . . . . . . . . . 18 3.1. AlignIO Classes diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 3.2. Align Classes diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 4.1. Blast Classes diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 4.2. BPLite Classes diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 4.3. Blast internal classes diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 4.4. Genscan Classes Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 5.1. Database Classes structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 6.1. UML meanings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 A.1. Bio::SeqIO structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59

List of Examples
2.1. 2.2. 2.3. 3.1. 3.2. 3.3. 4.1. 4.2. 4.3. 4.4. 4.5. 4.6. 4.7. 4.8. 4.9. 5.1. 5.2. SwissProt -> Fasta . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Loading a sequence from a remote server . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Find the references to the PDB database entries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Format conversions with AlignIO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 Basic methods of SimpleAlign . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 Filter gap columns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 StandAloneBlast run . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 StandAloneBlast parsing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 Parsing from a Blast le . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Parsing with BPLite . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Running PSI-blast . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 PSI-Blast SearchIO class . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Running bl2seq . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 Genscan parsing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Genscan parsing, with sub-sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Database class use . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 Database Index creation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

List of Exercises
2.1. Bio::SeqIO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2.2. An universal converter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2.3. Display a sequence in fasta format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 2.4. More on annotations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 2.5. Transmembran helices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 2.6. extractcds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 2.7. extractcds version 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 3.1. Create an alignment without gaps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 4.1. Running Blast on a Swissprot entry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 4.2. Running Blast: Setting parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 4.3. Running Blast: Saving output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 4.4. Running a Remote Blast . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 4.5. Display Blast hits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 4.6. Class of a Blast report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 4.7. Parse a Blast output le . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 4.8. Parse Blast results on standard input . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 4.9. Filtering hits by length . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 4.10. Filtering hits by position . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 4.11. Display the best hit by databank . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 4.12. Multiple queries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 4.13. Extracting the subject sequence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 4.14. Extracting alignments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 4.15. Locate EST in a genome . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 4.16. Record Blast hits as sequence features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 4.17. Print informations from a BPLite report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 4.18. Create a Bio::Tools::BPlite from a le . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 4.19. Create a Bio::Tools::BPlite from standard input . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 4.20. Parse a PSI-blast report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 4.21. Build a Bio::SimpleAlign object . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 4.22. Code reading: Bio::Tools::Genscan module . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 5.1. Parse a Genscan report and build a database entry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 5.2. Parse a Genscan report and build a database entry with the genomic sequence . . . . . . . . . . . . . . . . . . 47 5.3. Build a small bioperl module (for the golden program) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48

Chapter 1. General introduction

Chapter 1. General introduction

1.1. Bioperl documentation
1. Bioperl 1.0 tutorial. [http://bio.perl.org/Core/bptutorial.html] 2. Wiki Bioperl documentation. [http://bio.perl.org/wiki/html/BioPerl/FrontPage.html] 3. Modules listing. [http://doc.bioperl.org/releases/bioperl-1.0.1/] 4. http://doc.bioperl.org/bioperl-live/ [http://doc.bioperl.org/bioperl-live/] 5. Unix help: man and perldoc.

1.2. General bioperl classes

General bioperl classes [http://www.bioperl.org/images/bioperl.pdf] diagram (cf Figure 6.1)

Chapter 1. General introduction

Chapter 2. Sequences

Chapter 2. Sequences
2.1. The Bio::SeqIO class
2.1.1. Format Converter
Example 2.1. SwissProt -> Fasta
pseudocode: Construct a SwissProt formatted sequences input stream Construct a fasta formatted sequences output stream for all sequences: read the sequence from input stream write it to output stream in bioperl:
#! /local/bin/perl -w use strict; use Bio::SeqIO;

my $in

= Bio::SeqIO->newFh ( -file -format => swiss );

=> <seqs.sp,

my $out = Bio::SeqIO->newFh ( -file -format => fasta );

=> >seqs.fasta,

print $out $_ while <$in>;

datas: seqs.sp [data/seqs.sp]

Chapter 2. Sequences

Exercise 2.1. Bio::SeqIO

Find all available formats via the Bio::SeqIO [http://doc.bioperl.org/releases/bioperl-1.0.1/Bio/SeqIO.html] class (Solution A.1).

Exercise 2.2. An universal converter

Modify Example 2.1 to program an universal converter (Solution A.2).

2.2. Sequence classes

2.2.1. Introduction
Example 2.2. Loading a sequence from a remote server
pseudocode: Create a sequence object for the BACR_HALHA SwissProt entry Print its Accession number and description in bioperl:
use Bio::DB::SwissProt; my $database = new Bio::DB::SwissProt; my $seq = $database->get_Seq_by_id(BACR_HALHA); print "Seq: ", $seq->accession_number(), " -- ", $seq->desc(), "\n\n";

Chapter 2. Sequences

Exercise 2.3. Display a sequence in fasta format

Display the BACR_HALHA entry in fasta format (Solution A.3).

2.2.2. Building mechanisms summary

de novo:
use Bio::Seq; my $seq1 = Bio::Seq->new ( -seq => ATGAGTAGTAGTAAAGGTA, -id => my seq, -desc => this is a new Seq);

from a le:
use Bio::SeqIO; my $seqin = Bio::SeqIO->new ( -file => seq.fasta, -format => fasta); my $seq3 = $seqin->next_seq(); my $seqin2 = Bio::SeqIO->newFh ( -file -format my $seq4 = <$seqin2>; my $seqin3 = Bio::SeqIO->newFh ( -file -format my $seq5 = <$seqin3>; => seq.fasta, => fasta); => golden sp:TAUD_ECOLI |, => swiss);

by fetching an entry from a remote database:

use Bio::DB::SwissProt; my $banque = new Bio::DB::SwissProt; my $seq6 = $banque->get_Seq_by_id(KPY1_ECOLI);

by fetching an entry from a bioperl indexed local database:

use Bio::Index::Swissprot; my $inx = Bio::Index::Swissprot->new( -filename => small_swiss.inx); my $seq7 = $inx->fetch(100K_RAT);

Chapter 2. Sequences

data: small_swiss.inx [data/small_swiss.inx]

2.2.3. A deeper insight into the Bio::Seq class

Figure 2.1 shows the class structure of the Bio::Seq class. The Bio::Annotation package and Bio::SeqFeature package are more detailed in Figure 2.3 and Figure 2.5. Figure 2.2 shows the relation between a SwissProt entry and the corresponding Bio::Seq components.

Figure 2.1. Bio::Seq class structure

Chapter 2. Sequences

Chapter 2. Sequences

Figure 2.2. Relation of a SwissProt entry and the corresponding Bio::Seq object components

Chapter 2. Sequences

Figure 2.3. Bio::Annotation package structure

Example 2.3. Find the references to the PDB database entries

Does the protein have a known structure, and, if so, what are its cristallographic structures? Where can you nd this information within a SwissProt entry? Where can you nd this information in the sequence object?

Chapter 2. Sequences

in bioperl:
# PDB structures entries my @structures = (); foreach my $link ( $annotation->get_Annotations(dblink) ) { if ($link->database() eq PDB) { push (@structures, $link->primary_id()); } } print "\nPDB Structures: ", join (" ", @structures), "\n";

Exercise 2.4. More on annotations

Find the function of the protein within comments (if available) (Solution A.4).

Go to
Work on Exercise 2.6 before you continue.

Exercise 2.5. Transmembran helices

Find transmembrane helices in the entry (Solution A.5).

Tip
Use Figure 2.2 again. Use the appropriate objects documentation Here is a summary [codes/useSeq1.0.pl] solution of the exercises in this section.

2.2.4. Bioperl Sequence classes structure

10

Chapter 2. Sequences

Figure 2.4. Bioperl Sequence classes structure

2.3. Features and Location classes

2.3.1. Feature

11

Chapter 2. Sequences

They are created either by parsing database entries (see Bio::SeqIO::FTHelper [http://doc.bioperl.org/releases/bioperl1.0/Bio/SeqIO/FTHelper.html] or Figure A.1) or by parsing programs output. For instance (see Exercise 4.16), a Blast HSP is actually an instance of class Bio::SeqFeature::SimilarityPair [http://doc.bioperl.org/releases/bioperl-1.0/Bio/SeqFeature/SimilarityPair.html], which is a sub-class of Bio::SeqFeature::FeaturePair [http://doc.bioperl.org/releases/bioperl-1.0/Bio/SeqFeature/FeaturePair.html], which again is a feature. The same holds for Genscan predictions (Exercise 5.1). Documentation on features. Features table ofcial documentation [http://www.ebi.ac.uk/embl/Documentation/FT_denitions/feature_table.html]. bioperl tutorial [http://bio.perl.org/Core/bptutorial.html#III_7_1_Representing_sequence_an]. Features Classes. Bio::SeqFeatureI Bio::SeqFeature::Generic Bio::SeqFeature::FeaturePair Bio::SeqFeature::SimilarityPair Bio::SeqFeature::Similarity Bio::SeqFeature::Gene:: Bio::SeqFeature::Gene::ExonI Bio::SeqFeature::Gene::Exon Bio::SeqFeature::Gene::Transcript Bio::SeqFeature::Gene::TranscriptI Bio::SeqFeature::Gene::GeneStructureI Bio::SeqFeature::Gene::GeneStructure

12

Chapter 2. Sequences

Figure 2.5. Features Classes structure

2.3.2. Code reading: extracting CDS

Exercise 2.6. extractcds
code [lecture_code/extractcds] (download [ftp://ftp.pasteur.fr/pub/GenSoft/unix/nucleic_acid/extractcds] ) fetch an EMBL or Genbank entry and try this code with, for instance, gb:AB042770 (seq2.gb [data/seq2.gb]).

13

Chapter 2. Sequences

Web form [http://bioweb.pasteur.fr/seqanal/interfaces/extractcds.html]

Tip
Have a look at the Bio::SeqFeatureI [http://doc.bioperl.org/releases/bioperl1.0.1/Bio/SeqFeatureI.html] documentation. Figure 2.5 shows the architecture of the Bio::SeqFeature class.

Tip
Figure 2.6 shows the relation of features in an Embl entry and the Bio::SeqFeature class.

Exercise 2.7. extractcds version 2

compare Exercise 2.6 with this version [lecture_code/extractcds2]

2.3.3. Tag system

In the following example, CDS (feature key) is a primary tag, /codon a secondary tag.
CDS 1..1000 /codon=(seq:"cug",aa:Ser) /codon=(seq:"tga",aa:Trp)

This system enables you to refer to feature qualiers (see documentation [http://www.ebi.ac.uk/embl/Documentation/FT_denitions/featu

The tag system also enables to create new qualier types. Examples in this tutorial: Blast hits as features (see Figure 4.1 and Exercise 4.16). Creation of a database from a Genscan parsing (Exercise 5.1). See methods related to tags in Bio::SeqFeature::Generic 1.0/Bio/SeqFeature/Generic.html] documentation. [http://doc.bioperl.org/releases/bioperl-

14

Chapter 2. Sequences

Figure 2.6. Correspondance between an EMBL entry and bioperl tags

2.3.4. Location
Locations are useful for everything that looks like a range within a sequence (cf Bio::RangeI): sequences associated to features, sequences belonging to an alignment... Coordinates types. There are several coordinates types and representations: you can use fuzzy locations, ranges, lists, ... to specify the begin or the end (see the documentation

15

Chapter 2. Sequences

[http://www.ebi.ac.uk/embl/Documentation/FT_denitions/feature_table.html#location] ); example from Bio::Location::Fuzzy [http://doc.bioperl.org/releases/bioperl-1.0/Bio/Location/Fuzzy.html]:

my $fuzzylocation = new Bio::Location::Fuzzy(-start => <30, -end => 90, -loc_type => .);

taken

which species a location whose begin is before 30 and whose end is at 90. Coding: .. : EXACT, ^ : BETWEEN (a site between 2 bases), exemple : 123^124 a site between bases 123 and 124. 145^177 a site between 2 adjacent bases, somewhere between bases 145 and 177. . : WITHIN (a base within a range), example : (102.110) is a site wihtin the 102 .. 110 range, inclusive. < : BEFORE > : AFTER Location Classes. Bio::LocationI Bio::Location::Simple Bio::Location::SplitLocationIet Bio::Location::Split Bio::Location::FuzzyLocationIet Bio::Location::Fuzzy Coordinate policies, for fuzzy location start and end determination: interface: Bio::Location::CoordinatePolicyI, default: Bio::Location::WidestCoordPolicy Bio::Location::NarrowestCoordPolicy, Bio::Location::AvWithinCoordPolicy

16

Chapter 2. Sequences

Figure 2.7. Location Classes Structure

2.3.5. Graphical view of features

The version 1.0 of bioperl has a package to generate annotation images of sequences. The general procedure is to create a Bio::Graphics::Panel object that holds the image and the sequence to annotate. Then you can add annotations line per line by using the add_track method of the panel object. The Bio::Graphics::Glyph package contains several prdended styles to show annotations but you also have the possibility to write your own glyphs.

17

Chapter 2. Sequences

Figure 2.8 shows some features of the SwissProt entry BACR_HALHA. Here [codes/feature_graph.pl] is the that generates the image. The helix image and the arrow of domains are added glyph types that can be found here [modules/].

Figure 2.8. Graphical view of some features of the SwissProt entry BACR_HALHA

2.4. Sequence analysis tools

Bio::Seq [http://doc.bioperl.org/releases/bioperl-1.0.1/Bio/Seq.html] ( tutorial [http://bio.perl.org/Core/POD/bptutorial.html# :.

$seqobj->seq; # sequence as string $seqobj->subseq(10,40); # sub-sequence as string $seqobj->trunc(10,100); # sub-sequence as fresh Bio::PrimarySeq

18

Chapter 2. Sequences

$seqobj->translate;

Bio::Tools::SeqStats [http://doc.bioperl.org/releases/bioperl-1.0.1/Bio/Tools/SeqStats.html] ( tutorial [http://bio.perl.org/Core/POD/bptutorial.html#III_3_2_Obtaining_basic_sequence]) : .

$seq_stats = Bio::Tools::SeqStats->new(-seq=>$seqobj); $seq_stats->count_monomers(); $seq_stats->count_codons(); $weight = $seq_stats->get_mol_wt($seqobj);

et Bio::Tools::SeqWords [http:doc.bioperl.org/releases/bioperl-1.0.1/Bio/Tools/SeqWords.html//] :
$seq_word = Bio::Tools::SeqWords->new(-seq => $seqobj); $seq_stats->count_words($word_length);

Bio::Tools::SeqPattern [http://doc.bioperl.org/releases/bioperl-1.0.1/Bio/Tools/SeqPattern.html] ( tutorial [http://bio.perl.org/Core/POD/bptutorial.html#III_3_5_Miscellaneous_sequence_u]) : .

$pat = T[GA]AA...TAAT; $pattern = new Bio::Tools::SeqPattern(-SEQ =>$pat, -TYPE =>Dna); $pattern->expand; $pattern->revcom; $pattern->alphabet_ok;

Bio::SeqUtils [http://doc.bioperl.org/releases/bioperl-1.0.1/Bio/Tools/SeqUtils.html] : .
$util = new Bio::SeqUtils; $polypeptide_3char = $util->seq3($seqobj);

or:
$polypeptide_3char = Bio::SeqUtils->seq3($seqobj);

Bio::Tools::RestrictionEnzyme [http://doc.bioperl.org/releases/bioperl-1.0.1/Bio/Tools/RestrictionEnzyme.htm ( tutorial [http://bio.perl.org/Core/POD/bptutorial.html#III_3_3_Identifying_restriction_]) : .

$re1 = new Bio::Tools::RestrictionEnzyme(-NAME =>EcoRI); @fragments = $re1->cut_seq($seqobj); $re2 = new Bio::Tools::RestrictionEnzyme( -NAME =>EcoRV--GAT^ATC, # creates a new one

19

Chapter 2. Sequences

-MAKE =>custom);

Bio::Tools::Sigcleave [http://doc.bioperl.org/releases/bioperl-1.0.1/Bio/Tools/Sigcleave.html] ( tutorial [http://bio.perl.org/Core/POD/bptutorial.html#III_3_4_Identifying_amino_acid_c]) : .

$sigcleave_object = new Bio::Tools::Sigcleave (-file=>sigtest.aa, # not a Bio::Seq object! -threshold=>3.5, -desc=>test sigcleave protein seq, -type=>AMINO ); %raw_results = $sigcleave_object->signals; $formatted_output = $sigcleave_object->pretty_print;

20

Chapter 3. Alignments

Chapter 3. Alignments
3.1. AlignIO
The AlignIO class is designed the same as the SeqIO class. Therefore format conversions can be done as following:

Example 3.1. Format conversions with AlignIO

(data [data/inle.aln])
#! /local/bin/perl -w use strict; use Bio::AlignIO; my $inform = shift @ARGV || clustalw; my $outform = shift @ARGV || fasta; my $in = Bio::AlignIO->newFh ( -fh => \*STDIN, -format => $inform ); my $out = Bio::AlignIO->newFh ( -fh => \*STDOUT, -format => $outform ); print $out $_ while <$in>;

21

Chapter 3. Alignments

AlignIO class structure. Warning: some alignment formats are available only as input formats.

Figure 3.1. AlignIO Classes diagram

3.2. SimpleAlign
At rst, lets see some basic methods of the SimpleAlign class:

Example 3.2. Basic methods of SimpleAlign

#! /local/bin/perl -w use strict; use Bio::AlignIO; my $in = new Bio::AlignIO ( -file =>, $ARGV[0], -format => clustalw ); my $aln = $in->next_aln(); print "same length of all sequences: ", ($aln->is_flush()) ? "yes" : "no", "\n"; print "alignment lenght: ", $aln->length(), "\n"; printf "identity: %.2f %%\n", $aln->percentage_identity(); printf "identity of conserved columns: %.2f %%\n", $aln->overall_percentage_identity();

22

Chapter 3. Alignments

23

Chapter 3. Alignments

SimpleAlign class diagram. It contains the AlignI interface that declares all methods to be implemented by Alignment classes. The diagram integrates also some classes that can create SimpleAlign objects:

Figure 3.2. Align Classes diagram

The SimpleAlign class contains methods to select sequences or columns, but it can not lter alignments by functions as could be done by the UnivAln class of the old bioperl release. In order to lter columns by

24

Chapter 3. Alignments

properties, you have to extract the columns by yourself, lter them and reconstruct the new sequences. The following example lters gap columns.

Example 3.3. Filter gap columns

#! /local/bin/perl -w use strict; use Bio::AlignIO; my $in = new Bio::AlignIO ( -file =&gt; $ARGV[0], -format => clustalw ); my $out = newFh Bio::AlignIO ( -fh =&gt; \*STDOUT, -format => clustalw ); my $aln = $in->next_aln(); # if you need to working on the columns, create a list containing all columns as # strings my @aln_cols = (); foreach my $seq ( $aln->each_alphabetically() ) { my $colnr = 0; foreach my $chr ( split("", $seq->seq()) ) { $aln_cols[$colnr] .= $chr; $colnr++; } } # then do the work: # we want to eliminate all the columns containing gaps # 1/ we create a list containing all the columns without any gap my $gapchar = $aln-&gt;gap_char(); my @no_gap_cols = (); foreach my $col ( @aln_cols ) { next if $col =~ /\Q$gapchar\E/; push @no_gap_cols, $col; } # 2/ we modify the sequences in the alignment # 2a/ we reconstruct the sequence strings my @seq_strs = (); foreach my $col ( @no_gap_cols ) { my $colnr = 0; foreach my $chr ( split"", $col ) { $seq_strs[$colnr] .= $chr; $colnr++; }} # 2b/ we replace the old sequences strings with the new ones

25

Chapter 3. Alignments

foreach my $seq ( $aln->each_alphabetically() ) { $seq->seq(shift @seq_strs);} print $out $aln;

Exercise 3.1. Create an alignment without gaps

Create a new alignment without gaps at the begining and the end of the alignment. sample alignment le [data/inle.aln] Hint: you dont have to extract all the columns as in the second example Solution A.6

3.3. Code reading: protal2dna.

This script aligns DNA sequences, given their protein alignment. code [lecture_code/protal2dna.html] (download) bioperl classes: Bio::AlignIO [http://doc.bioperl.org/releases/bioperl-1.0/Bio/AlignIO.html] Bio::SimpleAlign [http://doc.bioperl.org/releases/bioperl-1.0/Bio/SimpleAlign.html] Bio::Tools::CodonTable [http://doc.bioperl.org/releases/bioperl-1.0/Bio/Tools/CodonTable.html] Web form [http://bioweb/seqanal/interfaces/protal2dna.html]

26

Chapter 4. Analysis

Chapter 4. Analysis
4.1. Blast
4.1.1. Running Blast
4.1.1.1. Running local Blast with Bio::Tools::Run::StandAloneBlast Example 4.1. StandAloneBlast run
use Bio::SeqIO; use Bio::Tools::Run::StandAloneBlast; my $Seq_in = Bio::SeqIO->new (-file => $ARGV[0], -format => fasta); my $query = $Seq_in->next_seq(); my $factory = Bio::Tools::Run::StandAloneBlast->new(program => blastp, database => swissprot, _READMETHOD => "Blast" ); my $blast_report = $factory->blastall($query); my $result = $blast_report->next_result; while( my $hit = $result->next_hit()) { print "\thit name: ", $hit->name(), " significance: ", $hit->significance(), "\n";}

You can try it with this le [data/1prot.fasta].

Exercise 4.1. Running Blast on a Swissprot entry

Run the above example with the TAUD_ECOLI entry from Swissprot (see Section 2.2.2) ). Solution A.7

Exercise 4.2. Running Blast: Setting parameters

Change the Expectation value (E) Blast parameter to 1.0 (instead of default value, 10.0) (query [data/1prot.fasta]). Solution A.8

Exercise 4.3. Running Blast: Saving output

Save the blast report in a local le (e.g blast.out). Solution A.9

27

Chapter 4. Analysis

Exercise 4.4. Running a Remote Blast

Find out from the bioperl Web site documentation how to run a Blast at NCBI. Solution A.10

4.1.2. Parsing Blast

4.1.2.1. Parsing from a Blast run result.
pseudocode: Run Blast. Create a report. For all hits (class Bio::Search::Hit::GenericHit [http://doc.bioperl.org/releases/bioperl-1.0/Bio/Search/Hit/GenericHit.html]) print its name and database for all HSP (class Bio::Search::HSP::GenericHSP 1.0/Bio/Search/HSP/GenericHSP.html]) print some hit statistics [http://doc.bioperl.org/releases/bioperl-

28

Chapter 4. Analysis

Example 4.2. StandAloneBlast parsing

use Bio::SeqIO; use Bio::Tools::Run::StandAloneBlast; my $Seq_in = Bio::SeqIO->new (-file => $ARGV[0], -format => fasta); my $query = $Seq_in->next_seq(); my $factory = Bio::Tools::Run::StandAloneBlast->new( program database => swissprot, _READMETHOD => "Blast" ); my $blast_report = $factory->blastall($query); my $result = $blast_report->next_result;

=> blastp,

while( my $hit = $result->next_hit()) { print "\thit name: ", $hit->name(), " significance: ", $hit->significance(), "\n"; while( my $hsp = $hit->next_hsp()) { print "E: ", $hsp->evalue(), "frac_identical: ", $hsp->frac_identical(), "\n";

}}

$result belongs to the Bio::Search::Result::GenericResult class, where you can nd the documentation about available methods on results.

Exercise 4.5. Display Blast hits

Display hit accession and hsp length. Solution A.11

Exercise 4.6. Class of a Blast report

What is the class of $blast_report? (hint: use the ref() perl function). Solution A.12

4.1.2.2. Parsing from a le.

The $blast_report that is created by the Bio::Tools::Run::StandAloneBlast new method actually belongs to the Bio::SearchIO class. So, you can directly create such an object from a le.

Example 4.3. Parsing from a Blast le

use Bio::SearchIO; my $blast_report = new Bio::SearchIO (-format => blast, -file => $ARGV[0]);

29

Chapter 4. Analysis

my $result = $blast_report->next_result; while( my $hit = $result->next_hit()) { print "\thit name: ", $hit->name(), "\n"; while( my $hsp = $hit->next_hsp()) { print "E: ", $hsp->evalue(), "frac_identical: ", $hsp->frac_identical(), "\n";

}}

Exercise 4.7. Parse a Blast output le

Run Example 4.3 with the le you have just created in Exercise 4.3.

Exercise 4.8. Parse Blast results on standard input

Run Example 4.3 with the blast report given on standard input. Solution A.13

4.1.2.3. Blast Classes diagram

30

Chapter 4. Analysis

Figure 4.1. Blast Classes diagram

4.1.2.4. Filtering Blast output Exercise 4.9. Filtering hits by length

Only display hits of length > to a given value (create a blast report from this le [data/blast.out]). Solution A.14

31

Chapter 4. Analysis

Exercise 4.10. Filtering hits by position

Only display hits between two positions. Solution A.15

Exercise 4.11. Display the best hit by databank

Display the best hit for each database in a Blast NCBI report (example [data/ncbi-blast.out]). If necessary, you can nd here a dbname(hit->name) [solution/dbname.pl] function to extract database name from the hit name. You may use a perl hash to mark a database as already done:
$done{$database} = 1;

Solution A.16

Exercise 4.12. Multiple queries

Submit two sequences and display the best hit for each. (2nd query [data/P42704.fasta]) Solution A.17

4.1.2.5. Extracting data from a Blast report Exercise 4.13. Extracting the subject sequence
Extract as a string the subject sequence from HSP with identity above a given level. Solution A.18

Exercise 4.14. Extracting alignments

Extract HSP alignment from hits which name contains a given pattern, and print this alignment in Clustalw format. Solution A.19

Exercise 4.15. Locate EST in a genome

Locate EST in a genome: displays the occurrences of this EST [data/est] in this contig [data/contig] of a genome; You may rst display this as a Clustalw alignment (on sequence by HSP); Blast report [data/blast-est.out] ; Solution A.20

32

Chapter 4. Analysis

Exercise 4.16. Record Blast hits as sequence features

Iterate a local Blast on databases given on the command line and record each corresponding best hit as a feature for the query sequence. Solution A.21

4.1.3. Bio::Tools::BPlite family parsers

The old Bio::Tools::BPlite [http://doc.bioperl.org/releases/bioperl-1.0/Bio/Tools/BPlite.html] family of parsers (BPlite [http://doc.bioperl.org/releases/bioperl-1.0/Bio/Tools/BPlite.html], Bio::Tools::BPpsilite [http://doc.bioperl.org/releases/bioperl-1.0/Bio/Tools/BPpsilite.html], Bio::Tools::BPbl2seq [http://doc.bioperl.org/releases/biop 1.0/Bio/Tools/BPbl2seq.html]) is still maintained in the 1.0 release, although is will be deprecated one day. It is actually still necessary for blasting two sequences and to perform a psiblast (Position Specic Iterative Blast). It is also still the default parser returned by Bio::Tools::Run::StandAloneBlast, as shown below (although this might change soon).

Example 4.4. Parsing with BPLite

In the following example (notice that we have removed the _READMETHOD parameter), the $blast_report does not belong to the Bio::SearchIO [http://doc.bioperl.org/releases/bioperl1.0/Bio/SearchIO.html] class, but to the Bio::Tools::BPlite [http://doc.bioperl.org/releases/bioperl1.0/Bio/Tools/BPlite.html] class. That is why the method to use in order to get informations are different, namely nextSbjct [http://doc.bioperl.org/releases/bioperl-1.0/Bio/Tools/BPlite.html#POD8] and nextHSP [http://doc.bioperl.org/releases/bioperl-1.0/Bio/Tools/BPlite/Sbjct.html#POD4] of class Bio::Tools::BPlite::Sbjct [http://doc.bioperl.org/releases/bioperl-1.0/Bio/Tools/BPlite/Sbjct.html] instead of next_hit and next_hsp. There is no next_result method, since these parsers were not able to deal with multiple reports.

use Bio::SeqIO; use Bio::Tools::Run::StandAloneBlast; my $Seq_in = Bio::SeqIO->new (-file => $ARGV[0], -format => fasta); my $query = $Seq_in->next_seq();; my $factory = Bio::Tools::Run::StandAloneBlast->new( program => blastp, database => swissprot ); my $blast_report = $factory->blastall($query); while (my $subject = $blast_report->nextSbjct()) { print $subject->name(), "\n"; while (my $hsp = $subject->nextHSP()) { print join("\t", $hsp->P, $hsp->percent, $hsp->score), "\n";

33

Chapter 4. Analysis

} }

Exercise 4.17. Print informations from a BPLite report

Take Example 4.4 and print hit names, and for all HSP, its P value, percent identity and score (see Bio::Tools::BPlite::HSP). Solution A.22

Exercise 4.18. Create a Bio::Tools::BPlite from a le

Create a Bio::Tools::BPlite from a le. Solution A.23

Exercise 4.19. Create a Bio::Tools::BPlite from standard input

Create a Bio::Tools::BPlite from standard input. Solution A.24

34

Chapter 4. Analysis

35

Chapter 4. Analysis

Figure 4.2. BPLite Classes diagram

36

Chapter 4. Analysis

4.1.4. PSI-BLAST (Position Specic Iterative Blast)

See documentation [http://www.ncbi.nlm.nih.gov/BLAST/tutorial/Altschul-2.html] or tutorial [http://www.ncbi.nlm.nih.gov/Edu at NCBI.

Example 4.5. Running PSI-blast

The following example run a PSI-blast:
use Bio::SeqIO; use Bio::Tools::Run::StandAloneBlast; my $query_in = Bio::SeqIO->newFh ( -file -format => fasta ); => $ARGV[0],

my $query = <$query_in>; $factory = Bio::Tools::Run::StandAloneBlast->new(database => swissprot); $factory->j(2); $blast_report = $factory->blastpgp($query); for my $iteration (1 .. $blast_report->number_of_iterations()) { print "iteration: $iteration\n"; my $result = $blast_report->round($iteration); while( my $hit = $result->nextSbjct()) { print "\thit name: ", $hit->name(), "\n"; foreach my $hsp ($hit->nextHSP) { print "\tstart: ", $hsp->hit->start(), " end: ",$hsp->hit->end(),"\n"; print "\tscore: ", $hsp->score(), "\n"; } }}

You can also load a report from a le, or from standard input:
my $blast_report = Bio::Tools::BPpsilite->new(-fh=>\*FH);

Example 4.6. PSI-Blast SearchIO class

There is also a SearchIO class for PSI-blast report, but it does not give access to the specic iterations:
use Bio::SearchIO; my $in = Bio::SearchIO->new( -format -file => $ARGV[0] ); => psiblast,

while ( my $result = $in->next_result() ) { foreach my $hit ( $result->hits ) {

37

Chapter 4. Analysis

print "Hit: $hit\n"; }}

Exercise 4.20. Parse a PSI-blast report

Load a PSI-blast report (example [data/psiblast.out] ) and at each iteration, display: the hits that were not present at the previous iteration the hits that have disappeared; Solution A.25

4.1.5. bl2seq: Blast 2 sequences

Example 4.7. Running bl2seq

use use use use use

Bio::SeqIO; Bio::Tools::Run::StandAloneBlast; Bio::AlignIO; Bio::SimpleAlign; Bio::LocatableSeq; => $ARGV[0],

my $query1_in = Bio::SeqIO->newFh ( -file -format => fasta ); my $query1 = <$query1_in>; my $query2_in = Bio::SeqIO->newFh ( -file -format => fasta ); my $query2 = <$query2_in>;

=> $ARGV[1],

my $out = Bio::AlignIO->newFh(-format => clustalw ); $factory = Bio::Tools::Run::StandAloneBlast->new( program => blastp ); $report = $factory->bl2seq($query1, $query2); while(my $hsp = $report->next_feature) { my $aln = Bio::SimpleAlign->new(); my $querySeq = Bio::LocatableSeq->new( -seq => $hsp->qs, -id => $query1->display_id,

38

Chapter 4. Analysis

-start => 1, -end => $query1->length ); my $sbjctSeq = Bio::LocatableSeq->new( -seq => $hsp->ss, -id => $report->sbjctName, -start => 1, -end => $query2->length ); $aln->add_seq($querySeq); $aln->add_seq($sbjctSeq); print $out $aln; }

Exercise 4.21. Build a Bio::SimpleAlign object

Build a Bio::SimpleAlign object from the subject and query sequences, and display this alignment in Clustalw format. Solution A.26

4.1.6. Blast Internal classes structure

39

Chapter 4. Analysis

The following diagram shows the internal classes structure. It is provided for information, and it has not to be understood to use bioperl Blast classes.

Figure 4.3. Blast internal classes diagram

40

Chapter 4. Analysis

4.2. Genscan
Example 4.8. Genscan parsing
The folowing example shows how to use the Bio::Tools::Genscan [http://doc.bioperl.org/releases/bioperl1.0/Bio/Tools/Genscan.html] parser. (data: Genscan output le [data/genscan.out])
use Bio::Tools::Genscan; my $genscan_file = $ARGV[0]; $genscan = Bio::Tools::Genscan->new(-file => $genscan_file); while(my $gene = $genscan->next_prediction()) { my $prot = $gene->predicted_protein; print "protein (", ref($prot), ") :\n", $prot->seq, "\n\n";}

Example 4.9. Genscan parsing, with sub-sequences

(data: Genscan output with CDS [data/genscan-cds.out])

# Genscan: example with sub-sequences use Bio::Tools::Genscan; my $genscan_file = $ARGV[0]; $genscan = Bio::Tools::Genscan->new(-file => $genscan_file); while(my $gene = $genscan->next_prediction()) { my $prot = $gene->predicted_protein; print "protein (", ref($prot), ") :\n", $prot->seq, "\n\n"; # display genscan predicted cds (if -cds genscan option) my $predicted_cds = $gene->predicted_cds; print "predicted CDS: \n", $predicted_cds->seq, "\n\n";

foreach my $exon ($gene->exons()) { my $loc = $exon->location; print "exon - primary_tag: ",$exon->primary_tag, " coding? ",$exon->is_coding(), " start-end: } foreach my $intron ($gene->introns()) { my $loc = $intron->location;

41

Chapter 4. Analysis

print "intron primary_tag: ",$intron->primary_tag, " start-end: ", $loc->start, "-", $loc->end, "\n" } foreach my $utr ($gene->utrs()) { my $loc = $utr->location; print "utr primary_tag: ",$utr->primary_tag, " start-end: ", $loc->start, "-", $loc->end, "\n"; } print "---------------------------------\n"; }

42

Chapter 4. Analysis

Figure 4.4. Genscan Classes Structure

43

Chapter 4. Analysis

Exercise 4.22. Code reading: Bio::Tools::Genscan module

1. What is the purpose of the Bio::SeqAnalysisParserI class? 2. Look at the _parse_predictions method in Bio::Tools::Genscan. 3. What happens during the creation of a Bio::Tools::Genscan instance? which methods are called? at which class hierarchy level?

44

Chapter 5. Databases

Chapter 5. Databases
5.1. Database classes
Example 5.1. Database class use
use Bio::Index::Swissprot; use Bio::SeqIO; my $out = Bio::SeqIO->newFh ( -fh => \*STDOUT, -format => fasta); my $Index_File_Name = shift; my $inx = Bio::Index::Swissprot->new( -filename => $Index_File_Name); foreach my $id (@ARGV) { my $seq = $inx->fetch($id); print $out $seq; }

Example 5.2. Database Index creation

(data: a SwissProt le to index [data/small_swiss.dat] ) :
use Bio::Index::Swissprot; my $Index_File_Name = shift; my $inx = Bio::Index::Swissprot->new( -filename => $Index_File_Name, -write_flag => 1); $inx->make_index(@ARGV);

45

Chapter 5. Databases

Figure 5.1. Database Classes structure

Exercise 5.1. Parse a Genscan report and build a database entry

Parse a Genscan report on a part of Arabidopsis chromosome V (data: Genscan report [data/genscan-arab.out]). (a) construct a database in Embl format

46

Chapter 5. Databases

containing an entry for each Genscan predicted CDS put this CDS exons as Features in the current entry (b) index this Embl formatted database (c) use your indexed database Outline of the exercise:

Solution A.27

47

Chapter 5. Databases

Exercise 5.2. Parse a Genscan report and build a database entry with the genomic sequence
Starting from exercise Exercise 5.1, add the DNA genomic sequence as sequence of the entry (+- delta at the beginning and at the end). Example:
SQ Sequence 2448 BP; 897 A; 487 C; 411 G; 653 T; 0 other; ttaagttcca ttgatgtatt tcaaagggtt cagagtttta tcgttttaca aagaaatgat gagtgttcat gactgtaaaa tccaccttca tcttccactt tcagtttaac ggctccggct

60 120

(data : Arabidopsis sequence - part of the chromosome V [data/arabidop-chr5-1quart.fasta]) build an additional feature describing all the CDS put the translation as a tag for this CDS (see add_tag_value [http://doc.bioperl.org/releases/bioperl1.0/Bio/SeqFeature/Generic.html#POD14]), which yields, for instance:
FT FT FT FT FT FT FT FT FT FT FT FT CDS join(complement(100..171),complement(284..358), complement(457..492),complement(626..673), complement(1460..1511),complement(2473..2513), complement(2638..2716),complement(2813..2969), complement(3046..3178),complement(3263..3390), complement(3476..3635)) /translation="MASTEGLMPITRAFLASYYDKYPFSPLSDDVSRLSSDMASLIKLL TVQSPPSQGETSLIDEANRQPPHKIDENMWKNREQMEEILFLLSPSRWPVQLREPSTSE DAEFASILRTLKDSFDNAFTAMISFQTKNSERIFSTVMTYMPQDFRGTLIRQQKERSER NKQAEVDALVSSGGSIRDTYALLWKQQMERRRQLAQLGSATGVYKTLVKYLVGVPQVLL DFIRQINDDDGDNEEYKEIIVQAGRTYEDIGFSVEYINASGEKTLILPYRRYEADQGNF STLMAGNYKLVWDNSYSTFFKKTLRYKVDCIAPVVEPDPEPEPLN"

regarding the location of the feature, see Section 2.3.4 in this tutorial, as well as add_sub_Location [http://doc.bioperl.org/releases/bioperl-1.0/Bio/Location/Split.html#POD2]. Solution A.28

5.2. Accessing a local database with golden

Exercise 5.3. Build a small bioperl module (for the golden program)
Idea: we want a bioperl module for local databases access. We could use bioperl indexes, except that this is not efcient for large databases (indexing time, and indexes size). All these databases are already indexed by golden [ftp://ftp.pasteur.fr/pub/GenSoft/unix/db_soft/golden] in a very efcient way. The module could be designed to be used like this:

48

Chapter 5. Databases

use LocalDBAccess; use Bio::SeqIO; my $seq = LocalDBAccess->get_Seq_by_id ( $ARGV[0] ); my $out = Bio::SeqIO->newFh ( -fh => \*STDOUT, -format => fasta );

print $out $seq;

Solution A.29

49

Chapter 5. Databases

50

Chapter 6. Perl Reminders

Chapter 6. Perl Reminders

These reminders may be helpful to use modules [#use], or to a more advanced understanding [#advanced] of them.

6.1. UML
Figure 6.1. UML meanings

6.2. Perl reminders to use bioperl modules

6.2.1. References
to pass a hash or a function as a parameter to a function:
$blastObj = Bio::Tools::Blast->new( -run -parse -filt_func => \&my_filter); => \%runParam, => 1,

for an output parameter:

%blastParam = (-run => \%runParam, -save_array => \@blast_objs );

51

Chapter 6. Perl Reminders

to pass a lehandle as a parameter:

my $banque = Bio::SeqIO->newFh ( -fh -format => embl); => \*STDOUT,

references on an anonymous array:

%runParam = ( -method -prog -database -seqs => => => => remote, blastp, swissprot, [ $seq ]); # Bio::Seq.pm objects.

which is "equivalent" to:

%runParam = ( -method -prog -database -seqs => => => => remote, blastp, swissprot, \@seqs); # Bio::Seq.pm objects.

with a variable. references on anonymous data structures:

# (reminder) anonymous hash: %a = (a => 1, b => 2 ); print "hash ", ref(%a), " ", $a{a}, " ", $a{b}, "\n"; # reference on an anonymous hash: $ra = {a => 1, b => 2 }; print "ref hash ", ref($ra), " ", $ra->{a}, " ", $ra->{b}, "\n"; # reference on an anonymous array: $rl = [1, 2]; print "ref array ", ref($rl), " ", $rl->[0], " ", $rl->[1], "\n";

See for instance in exercises: 2.4 et 2.7. ref() returns the type of the variable (or its class,see below).

6.2.2. Filehandles and streams

A data stream is for instance an open le. Streams are available through "lehandles" (le descriptors) - in perl: a symbol without a $, such as INFILE or OUTFILE. Standard input and standard output streams are associated by default to the STDIN and STDOUT descriptors. The lehandle is associated to the stream when opening the le:

52

Chapter 6. Perl Reminders

open (INFILE, $infile) || die "cannot open $infile:$!"; open (OUTFILE, "> $outfile") || die "cannot open $outfile:$!"; $line = <INFILE> ; # read a sequence objectprint OUTFILE $sequence; # write a sequence object

In bioperl, there are classes that build lehandles:

$in = Bio::SeqIO->newFh ( -file => f) ; # make a tied filehandle $sequence = <$in>; # read a sequence object $out = Bio::SeqIO->newFh ( -file => > f); print $fh $sequence; # write a sequence object

Filhandles such as INFILE or STDOUT belong to a specic namespace and can not be used in an assignation or as a parameter. For this, you have to pass a reference to the typeglob:
my $banque = Bio::SeqIO->newFh ( -fh => \*STDOUT, -format => embl);

typeglobs are in a general namespace, on top of the other namespaces for scalars, arrays, hashes and functions. You can use it to create aliases (see the "Advanced Perl Programming" book).

6.2.3. Exceptions
When using a bioperl module, the following can happen:

-------------------- EXCEPTION -------------------MSG: Attempting to set the sequence to [==] which does not look healthy STACK Bio::PrimarySeq::seq /local/lib/perl5/site_perl/5.6.0/Bio/PrimarySeq.pm:243 STACK Bio::PrimarySeq::new /local/lib/perl5/site_perl/5.6.0/Bio/PrimarySeq.pm:218 STACK Bio::Seq::new /local/lib/perl5/site_perl/5.6.0/Bio/Seq.pm:132 STACK Bio::SeqIO::fasta::next_primary_seq /local/lib/perl5/site_perl/5.6.0/Bio/SeqIO/fasta.p STACK Bio::SeqIO::fasta::next_seq /local/lib/perl5/site_perl/5.6.0/Bio/SeqIO/fasta.pm:85 STACK toplevel solution/blast_features.pl:22 -------------------------------------------

bioperl indeed use perl exception mechanisms (die et eval) to handle use errors. An exception is raised by the perl module an can be caught:

53

Chapter 6. Perl Reminders

throw : in bioperl, this is the method for raising exceptions. It is provided by the Bio::Root::RootI [http://doc.bioperl.org/releases/bioperl-1.0/Bio//Root/RootI.html] class:

sub throw{ my ($self,$string) = @_; my $std = $self->stack_trace_dump(); my $out = "-------------------- EXCEPTION --------------------\n"."MSG: ".$string."\n".$std."--die $out;}

You can catch exceptions with eval (dont forget the ; at the end of the eval block:
eval { # this instruction may throw an exception # if parameters are incorrect $report = $factory->bl2seq($querySeq, $subjectSeq ); }; if( $@ ) { print "Caught exception"; } else { print "no exception"; }

In bioperl, exceptions are also used to implement interfaces classes; the following example in Bio::PrimarySeqI [http://doc.bioperl.org/releases/bioperl-1.0/Bio/PrimarySeqI.html]:

sub seq { my ($self) = @_; if( $self->can(throw) ) { $self->throw("Bio::PrimarySeqI definition of seq - implementing class did not provide this met

} else { confess("Bio::PrimarySeqI definition of seq - implementing class did not provide this method") } }

is a mean to oblige the subclass to implement a seq method. It is thus a way to encode abstract methods.

54

Chapter 6. Perl Reminders

6.2.4. Getopt::Std
To handle command line options: man Getopt::Std. Example :
use Getopt::Std; my %opts = ();getopts(i:o:I:O:, \%opts); my $infile = $opts{I} || infile; my $outfile = $opts{O} || outfile; my $inform = $opts{i} || swiss; my $outform = $opts{o} || fasta;

Another example :
use Getopt::Std;getopts(f:plgh); if ($opt_f) { $format = $opt_f;} else { $format = "Genbank";}

6.2.5. Classes
Inheritance, example in Bio::Seq [http://doc.bioperl.org/releases/bioperl-1.0/Bio/Seq.html]:
@ISA = qw(Bio::Root::RootI Bio::SeqI);

In bioperl, you use classes the following ways: By instantiating a class, most often with a new method (this is a coding convention - see also newFh [http://doc.bioperl.org/releases/bioperl-1.0/Bio/SeqIO.html#newFh], from_searchResult [http://doc.bioperl.org/releases/bioperl-1.0/Bio/SeqFeature/SimilarityPair.html#from_searchResult], ...) :
$seq_stats = Bio::Tools::SeqStats->new ($seqobj);

As a library component.

55

Chapter 6. Perl Reminders

Test whether a class or an object is-a:

if (Bio::Tools::Blast->isa(Bio::Root::RootI)) { ...} if ($seq->isa(Bio::Seq::RichSeq)) { ...}

nd the class of an object:

print "class of exon: ", ref($exon), "\n";

6.2.6. BEGIN block

Statements to be executed when the module is loaded (e.g when issuing a use Module; statement) ; see for instance: Bio::LocationI to dene the default coordinate policy:
BEGIN { $coord_policy = Bio::Location::WidestCoordPolicy->new();}

Remark: this is necessary in object-oriented style, since everything is encapsulated in methods.

6.3. Perl reminders for a further advanced understanding of bioperl modules

6.3.1. Modules
Exporter [http://www.perldoc.com/perl5.6/lib/Exporter.html], handles module variables scope (even though there are no private variables in perl). @EXPORT= qw(...); : symbols exported by default; @EXPORT_OK = qw(...); : symbols imported on demand;

56

Chapter 6. Perl Reminders

%EXPORT_TAGS = tag => [...]; tags for symbols sets; example :

use Bio::Tools::Blast qw(:obj);

imports symbols tagged as obj - here:$Blast, a static Bio::Tools::Blast object for a restrcited use of the module methods.

6.3.2. Compiler instructions

use strict: no symbolic references, no access to non local variables (local variable are declared by a my statement), no barewords. use vars qw(@ISA %valid_type); : a "pragma"to pre-declare global variables.

6.3.3. Tie
: Associates a class to a variable. whenever a variable is "tied",read and write statements (access, assignation, ...) trigger a call topredened subroutines (FETCH, STORE, PRINT, READLINE...). You can nd an example in (Bio::SeqIO):
sub fh { my $self = shift; my $class = ref($self) || $self; my $s = Symbol::gensym; tie $$s,$class,$self; return $s; }

which returns a lehandle tied to the Bio::SeqIO::Fhclass.As explained in Tying-FileHandles, Bio::SeqIO class has thus to redene these methods:
sub READLINE { my $self = shift; return $self->{seqio}->next_seq() unless wantarray; my (@list, $obj); push @list, $obj while $obj = $self->{seqio}->next_seq(); return @list; } sub PRINT { my $self = shift; }

$self->{seqio}->write_seq(@_);

These methods enable the programmer to read and print through the variable:

57

Chapter 6. Perl Reminders

$fh = $obj->fh; # make a tied filehandle$sequence = <$fh>; # read a sequence object (calls READLINE) print $fh $sequence; # write a sequence object (calls PRINT)

58

Appendix A. Solutions

Appendix A. Solutions
A.1. Sequences
Solution A.1. Bio::SeqIO structure
Exercise 2.1

59

Appendix A. Solutions

Figure A.1. Bio::SeqIO structure

Solution A.2. An universal converter

Exercise 2.2 UnivConvert via a le.

60

Appendix A. Solutions

#! /local/bin/perl -w # exercise 1.1 : UnivConvert via file use strict; use Bio::SeqIO; my $inform = shift @ARGV; my $outform = shift @ARGV; my $infile = shift @ARGV; my $outfile = shift @ARGV; if ( ! defined $outfile ) { $outfile = $infile; $outfile =~ s/\..*$// if $outfile =~ /\./; $outfile .= "." . $outform; } my $in = Bio::SeqIO->newFh ( -file -format => $inform ); => $infile,

my $out = Bio::SeqIO->newFh ( -file -format => $outform ); print $out $_ while <$in>;

=> ">$outfile",

You can use it like this:

perl UnivConvert2.pl swiss gcg seqs.sp

UnivConvert via stream.

#! /local/bin/perl -w # exercise 1.1 : UnivConvert via flux use strict; use Bio::SeqIO; my $inform = shift @ARGV || swiss; my $outform = shift @ARGV || fasta; my $in = Bio::SeqIO->newFh ( -fh -format => $inform ); => \*STDIN,

my $out = Bio::SeqIO->newFh ( -fh

=> \*STDOUT,

61

Appendix A. Solutions

-format => $outform );

print $out $_ while <$in>;

You can use it like this:

cat seqs.sp | perl ../solution/UnivConvertStdmin.pl swiss gcg

UnivConvert with options.

#! /local/bin/perl -w # exercice 1.1 : UnivConvert avec options use strict; use Getopt::Std; use Bio::SeqIO; my %opts = (); getopts(i:o:I:O:, \%opts); my my my my $infile $outfile $inform $outform = = = = $opts{I} $opts{O} $opts{i} $opts{o} || || || || infile; outfile; swiss; fasta;

my $in

= Bio::SeqIO->newFh ( -file -format => $inform );

=> $infile,

my $out = Bio::SeqIO->newFh ( -file -format => $outform );

=> ">$outfile",

print $out $_ while <$in>;

You can use it like this:

perl ../solution/UnivConvert.pl -I seqs.sp -O seqs.fasta

UnivConvert with options and stream.

#! /local/bin/perl -w

62

Appendix A. Solutions

# exercise 1.1 : UnivConvert with option and stream use strict; use Getopt::Std; use Bio::SeqIO; my %opts = (); getopts(hi:o:, \%opts); usage() if (exists $opts{h}); my $inform = lc($opts{i}) || swiss; my $outform = lc($opts{o}) || fasta; (format_ok($inform) && format_ok($outform)) || exit 1; my $in = Bio::SeqIO->newFh ( -fh -format => $inform ); => \*STDIN,

my $out = Bio::SeqIO->newFh ( -fh -format => $outform );

=> \*STDOUT,

print $out $_ while <$in>; sub format_ok { my $form = shift @_; my %formats = qw(fasta 1 swiss 1 embl 1 genbank 1 scf 1 pir 1 gcg 1 raw 1 ace 1 bsml 1 fastq 1 phd 1 qual 1); if (! exists $formats{$form}) { print STDERR $form, " -- unknown format\n"; usage (); return 0; # false } return 1; # true }

sub usage { my $prog =basename $0; chomp($prog); print "\n"; print "$prog convert files from one format into another format\n"; print "\n";

63

Appendix A. Solutions

print print print print print print print print

"$prog [-i informat] [-o outformat] < infile\n"; "\n"; "-i informat -- infile format (default swiss)\n"; "-o outformat -- outfile format (default fasta)\n"; "\n"; "-h -- print this message\n"; "\n"; "Available formats: fasta, swiss, EMBL, GenBank, SCF, Pir, GCG, raw et ACE\n";

print "\n"; exit 1; }

You can use it like this:

cat seqs.sp | perl ../solution/UnivConvertStd.pl -o gcg > seqs.gcg

Solution A.3. Display a sequence in fasta format

Exercise 2.3
# rest de lannotation my $annotation &equals; $seq-&gt;annotation(); my @structures &equals; (); foreach my $link ( $annotation-&gt;each_DBLink() ) { if ($link-&gt;database() eq PDB) { push (@structures, $link-&gt;primary_id()); } }

Solution A.4. More on annotations

Exercise 2.4
# get the protein function -- exo my $function = ""; foreach my $comment ( $annotation->get_Annotations(comment) ) { my @lines = split (/\.\n/, $comment->text()); foreach my $line (@lines) { if ( $line =~ s/^-!- FUNCTION:// ) { $line =~ s/\n/ /g; $function = $line;

64

Appendix A. Solutions

last; } } last if $function ne ""; } print "\nFunction:", $function, "\n";

Solution A.5. Transmembran helices

Exercise 2.5
# almost for GenBank/EMBL entries my @features = $seq->top_SeqFeatures(); foreach my $feat (@features) { if ($feat->primary_tag() eq TRANSMEM) { print "TM Segment: from ", $feat->start(), " to ", $feat->end(), "\n"; } }

A.2. Alignments
Solution A.6. Create an alignment without gaps
Exercise 3.1
#! /local/bin/perl -w ### ### Create a new alignment without gaps at the begining and the end of an alignment

use strict; use Bio::AlignIO; # get alignment my $in; if ( $#ARGV > -1 ) { $in = new Bio::AlignIO ( -file =>, $ARGV[0], -format => clustalw ); } else { $in = new Bio::AlignIO ( -fh => \*STDIN, -format => clustalw ); } my $out = newFh Bio::AlignIO ( -fh => \*STDOUT, -format => clustalw );

65

Appendix A. Solutions

my $aln = $in->next_aln(); # find max column index of starting gaps and # min column index of ending gaps my $startgaps = 0; my $endgaps = 0; my $gap_char = $aln->gap_char(); foreach my $seq ($aln->each_seq) { my $str = $seq->seq(); $str =~/^(\Q$gap_char\E*)/; my $len = length($1); $startgaps = ($startgaps > $len) ? $startgaps : $len; $str =~/(-*)$/; $len = length($1); $endgaps = ($endgaps > $len) ? $endgaps : $len; } # cut the starting and ending block containing gaps print $out $aln->slice($startgaps+1, $aln->length()-$endgaps);

A.3. Analysis
Solution A.7. Running Blast on a Swissprot entry
Exercise 4.1
use Bio::SeqIO; use Bio::Tools::Run::StandAloneBlast; # a) solution if you have a local program to fetch # database entries # (ftp://ftp.pasteur.fr/pub/GenSoft/unix/db_soft/golden/) #my $Seq_in = Bio::SeqIO->new (-file => golden sp:TAUD_ECOLI |, # -format => swiss); #my $query = $Seq_in->next_seq();; # b) else: use Bio::DB::SwissProt; my $database = new Bio::DB::SwissProt; my $query = $database->get_Seq_by_id(TAUD_ECOLI); my $factory = Bio::Tools::Run::StandAloneBlast->new(

66

Appendix A. Solutions

program => blastp, database => swissprot, _READMETHOD => "Blast" ); my $blast_report = $factory->blastall($query); my $result = $blast_report->next_result; while( my $hit = $result->next_hit()) { print "\thit name: ", $hit->name(), " significance: ", $hit->significance(), "\n"; }

Solution A.8. Running Blast: Setting parameters

Exercise 4.2
use Bio::SeqIO; use Bio::Tools::Run::StandAloneBlast; my $Seq_in = Bio::SeqIO->new (-file => $ARGV[0], -format => fasta); my $query = $Seq_in->next_seq();; my $factory = Bio::Tools::Run::StandAloneBlast->new( program => blastp, database => swissprot, _READMETHOD => "Blast" ); $factory->e(1.0); my $blast_report = $factory->blastall($query); my $result = $blast_report->next_result; while( my $hit = $result->next_hit()) { print "\thit name: ", $hit->name(), " significance: ", $hit->significance(), "\n"; }

Solution A.9. Running Blast: Saving output

Exercise 4.3
use Bio::SeqIO; use Bio::Tools::Run::StandAloneBlast; my $Seq_in = Bio::SeqIO->new (-file => $ARGV[0], -format => fasta); my $query = $Seq_in->next_seq();; my $factory = Bio::Tools::Run::StandAloneBlast->new( program => blastp,

67

Appendix A. Solutions

database => swissprot, _READMETHOD => "Blast" ); $factory->outfile(blast.out); my $blast_report = $factory->blastall($query); my $result = $blast_report->next_result; while( my $hit = $result->next_hit()) { print "\thit name: ", $hit->name(), " significance: ", $hit->significance(), "\n"; }

Solution A.10. Running a remote Blast

Exercise 4.4
use Bio::SeqIO; use Bio::Tools::Run::RemoteBlast; my $Seq_in = Bio::SeqIO->new (-file => $ARGV[0], -format => fasta); my $query = $Seq_in->next_seq(); my $factory = Bio::Tools::Run::RemoteBlast->new( -prog => blastp, -data => swissprot, _READMETHOD => "Blast" ); my $blast_report = $factory->submit_blast($query);

while ( my @rids = $factory->each_rid ) { print STDERR "waiting...\n" ; # RID = Remote Blast ID (e.g: 1017772174-16400-6638) foreach my $rid ( @rids ) { my $rc = $factory->retrieve_blast($rid); if( !ref($rc) ) { if( $rc < 0 ) { # retrieve_blast returns -1 on error $factory->remove_rid($rid); } # retrieve_blast returns 0 on job not finished sleep 5; } else { #---- Blast done ---$factory->remove_rid($rid);

68

Appendix A. Solutions

my $result = $rc->next_result; print "database: ", $result->database_name(), "\n"; while( my $hit = $result->next_hit ) { print "hit name is: ", $hit->name, "\n"; while( my $hsp = $hit->next_hsp ) { print "score is: ", $hsp->score, "\n"; } } } } }

Solution A.11. Display Blast hits

Exercise 4.5
use Bio::SeqIO; use Bio::Tools::Run::StandAloneBlast; my $Seq_in = Bio::SeqIO->new (-file => $ARGV[0], -format => fasta); my $query = $Seq_in->next_seq();; my $factory = Bio::Tools::Run::StandAloneBlast->new( program => blastp, database => swissprot, _READMETHOD => "Blast" ); my $blast_report = $factory->blastall($query); my $result = $blast_report->next_result; while( my $hit = $result->next_hit()) { print "\thit accession: ", $hit->accession(), "\n"; while( my $hsp = $hit->next_hsp()) { print "length: ", $hsp->length(), "\n"; } }

Solution A.12. Class of a Blast report

Exercise 4.6
use Bio::SeqIO; use Bio::Tools::Run::StandAloneBlast; my $Seq_in = Bio::SeqIO->new (-file => $ARGV[0], -format => fasta); my $query = $Seq_in->next_seq();;

69

Appendix A. Solutions

my $factory = Bio::Tools::Run::StandAloneBlast->new( program => blastp, database => swissprot, _READMETHOD => "Blast" ); my $blast_report = $factory->blastall($query); print "Class of blast_report is: ", ref($blast_report), "\n";

Solution A.13. Parse Blast results on standard input

Exercise 4.8
use Bio::SearchIO; my $blast_report = new Bio::SearchIO (-format => blast, -fh => \*STDIN ); my $result = $blast_report->next_result; while( my $hit = $result->next_hit()) { print "\thit name: ", $hit->name(), "\n"; while( my $hsp = $hit->next_hsp()) { print "E: ", $hsp->evalue(), "frac_identical: ", $hsp->frac_identical(), "\n"; } }

One way to use this code is by feeding a blastall output directly to the script through a Unix pipe:
blastall -p blastp -d swissprot -i data/1prot.fasta | perl solution/blast_parse_stdin.pl

Solution A.14. Filtering hits by length

Exercise 4.9
use Bio::SearchIO; my $max_length = ($ARGV[1])? $ARGV[1] : 200; my $blast_report = new Bio::SearchIO (-format => blast, -file => $ARGV[0]); my $result = $blast_report->next_result; while( my $hit = $result->next_hit()) { if ($hit->length() >= $max_length) { print "\thit name: ", $hit->name(), " hit length: ", $hit->length(), "\n"; while( my $hsp = $hit->next_hsp()) {

70

Appendix A. Solutions

print "E: ", $hsp->evalue(), "frac_identical: ", $hsp->frac_identical(), "\n"; } } }

Solution A.15. Filtering hits by position

Exercise 4.10
use Bio::SearchIO; my $min_start = ($ARGV[1])? $ARGV[1] : 1; my $max_end = ($ARGV[2])? $ARGV[2] : -1; my $blast_report = new Bio::SearchIO (-format => blast, -file => $ARGV[0]); my $result = $blast_report->next_result; while( my $hit = $result->next_hit()) { print "\thit name: ", $hit->name(), "\n"; while( my $hsp = $hit->next_hsp()) { if ($hsp->start() >= $min_start && ($max_end == -1 || ($hsp->end() <= $max_end)) ) { print "\t\tstart: ", $hsp->start(), " end: ", $hsp->end(), "\n"; print "\t\tE: ", $hsp->evalue(), "frac_identical: ", $hsp->frac_identical(), "\n"; } } }

Solution A.16. Display the best hit by databank

Exercise 4.11
use Bio::SearchIO; sub dbname { my ($name) = @_; $db= $name; $db =~ /(\w+)|.*/; $db = $1; return $db; } my $blast_report = new Bio::SearchIO (-format => blast,

71

Appendix A. Solutions

-file => $ARGV[0]); my $result = $blast_report->next_result; while( my $hit = $result->next_hit()) { my $dbname = dbname($hit->name()); if ($done{$dbname}) { next; } else { $done{$dbname} = 1; } print "\thit name: ", $hit->name(), "\n"; while( my $hsp = $hit->next_hsp()) { print "\t\tE: ", $hsp->evalue(), "frac_identical: ", $hsp->frac_identical(), "\n"; } }

Solution A.17. Multiple queries

Exercise 4.12
use Bio::SeqIO; use Bio::Tools::Run::StandAloneBlast; my @query_seqs = (); my $Seq1_in = Bio::SeqIO->new (-file => $ARGV[0], -format => fasta); push (@query_seqs, $Seq1_in->next_seq()); my $Seq2_in = Bio::SeqIO->new (-file => $ARGV[1], -format => fasta); push (@query_seqs, $Seq2_in->next_seq()); my $factory = Bio::Tools::Run::StandAloneBlast->new( program => blastp, database => swissprot, _READMETHOD => "Blast" ); my $blast_report = $factory->blastall(\@query_seqs); while (my $result = $blast_report->next_result()) { while( my $hit = $result->next_hit()) { print "\thit name: ", $hit->name(), " significance: ", $hit->significance(), "\n"; last; } }

72

Appendix A. Solutions

Solution A.18. Extracting the subject sequence

Exercise 4.13
use Bio::SearchIO; my $blast_report = new Bio::SearchIO (-format => blast, -file => $ARGV[0]); my $result = $blast_report->next_result; my $level = $ARGV[1]; while( my $hit = $result->next_hit()) { while( my $hsp = $hit->next_hsp()) { if ($hsp->frac_identical() >= $level) { print $hsp->hit_string, "\n"; } } }

Solution A.19. Extracting alignments

Exercise 4.14
use Bio::SearchIO; my $blast_report = new Bio::SearchIO (-format => blast, -file => $ARGV[0]); my $result = $blast_report->next_result; my $level = $ARGV[1]; while( my $hit = $result->next_hit()) { while( my $hsp = $hit->next_hsp()) { if ($hsp->frac_identical() >= $level) { print $hsp->hit_string, "\n"; } } }

Solution A.20. Locate EST in a genome

Exercise 4.15
use use use use use Bio::SeqIO; Bio::SearchIO; Bio::AlignIO; Bio::SimpleAlign; Bio::LocatableSeq;

73

Appendix A. Solutions

my $seq_in = Bio::SeqIO->new (-file => $ARGV[0], -format => fasta); my $seq = $seq_in->next_seq(); my $contig_seq = Bio::LocatableSeq->new( -seq => $seq->seq, -id => $seq->display_id, -start => 1, -end => $seq->length ); my $aln = Bio::SimpleAlign->new(); $aln->add_seq($contig_seq); my $blast_report = new Bio::SearchIO (-format => blast, -file => $ARGV[1]); my $result = $blast_report->next_result; while( my $hit = $result->next_hit()) { my $i = 0; while( my $hsp = $hit->next_hsp()) { $i++; my $start = $hsp->hit->start(); my $end = $hsp->hit->end(); my $name = $result->query_name . "_HSP_$i"; my $seq = fill_gaps($hsp->query_string, $start, $aln->length); my $query_seq = Bio::LocatableSeq->new( -seq => $seq, -id => $name, -start => $start, -end => $end ); $aln->add_seq($query_seq); } } my $out = Bio::AlignIO->newFh(-format => clustalw ); print $out $aln; sub fill_gaps { # add gaps from 0 to start and from start + seq length to end my ($seq, $start, $length) = @_; my $gapped_seq = ""; for (my $i = 0; $i < $start - 1; $i++) { $gapped_seq .= "-"; } $gapped_seq .= $seq; for (my $i = $start + length($seq); $i <= $length; $i++) { $gapped_seq .= "-"; } return $gapped_seq; }

74

Appendix A. Solutions

Example of use:
perl solution/blast_locate.pl data/contig data/blast-est.out

Solution A.21. Record Blast hits as sequence features

Exercise 4.16
use use use use my my my my Bio::SeqIO; Bio::Tools::Run::StandAloneBlast; Bio::Seq; Bio::SeqFeature::SimilarityPair; $seqfile = shift @ARGV; $in = Bio::SeqIO->new (-file => $seqfile, -format => fasta); $query = $in->next_seq(); @dbs = @ARGV;

foreach my $db (@dbs) { my $factory = Bio::Tools::Run::StandAloneBlast->new(database => $db, program => blastp); my $report = $factory->blastall($query); while(my $sbjct = $report->nextSbjct) { print STDERR "name : ", $sbjct->name, "\n"; while (my $hsp = $sbjct->nextHSP) { print STDERR "GFF:\n", $hsp->query()->gff_string(), "\n"; $query->add_SeqFeature($hsp); } last; } } foreach my $feat ($query->all_SeqFeatures()) { print STDERR "Feature from : ", $feat->start, " to : ", $feat->end, " Primary tag : ", $feat->primary_tag, ", produced by ", $feat->source_tag(), "\n"; $feat->primary_tag("BLAST"); } $out = Bio::SeqIO->newFh(-fh => \*STDOUT , -format => swiss); print $out $query;

75

Appendix A. Solutions

Solution A.22. Print informations from a BPLite report

Exercise 4.17
use Bio::SeqIO; use Bio::Tools::Run::StandAloneBlast; my $Seq_in = Bio::SeqIO->new (-file => $ARGV[0], -format => fasta); my $query = $Seq_in->next_seq();; my $factory = Bio::Tools::Run::StandAloneBlast->new( program => blastp, database => swissprot ); my $blast_report = $factory->blastall($query); while (my $subject = $blast_report->nextSbjct()) { print $subject->name(), "\n"; while (my $hsp = $subject->nextHSP()) { print join("\t", $hsp->P, $hsp->percent, $hsp->score), "\n"; } }

Solution A.23. Create a Bio::Tools::BPlite from a le

Exercise 4.18
use Bio::Tools::BPlite; my $report = new Bio::Tools::BPlite(-file => $ARGV[0]); while (my $subject = $report->nextSbjct()) { print $subject->name(), "\n"; while (my $hsp = $subject->nextHSP()) { print join("\t", $hsp->P, $hsp->score, $hsp->length, $hsp->match, $hsp->positive), "\n"; } }

76

Appendix A. Solutions

Solution A.24. Create a Bio::Tools::BPlite from standard input

Exercise 4.19
use Bio::Tools::BPlite; my $report = new Bio::Tools::BPlite(-fh => \*STDIN); while (my $subject = $report->nextSbjct()) { print $subject->name(), "\n"; while (my $hsp = $subject->nextHSP()) { print join("\t", $hsp->P, $hsp->score, $hsp->length, $hsp->match, $hsp->positive), "\n"; } }

Solution A.25. Parse a PSI-blast report

Exercise 4.20
use Bio::Tools::BPpsilite; my $blast_report = new Bio::Tools::BPpsilite (-file => $ARGV[0]);

for my $iteration (1 .. $blast_report->number_of_iterations()) { print "\n-----------------------------\niteration: $iteration\n"; my $result = $blast_report->round($iteration); my $hitarray_ref = $result->newhits; foreach my $hit (@{ $hitarray_ref }) { print "NEW hit name: $hit\n"; } my $oldhitarray_ref = $result->oldhits; foreach my $hit (@previous_hitarray) { if (grep /\Q$hit\E/, @{ $oldhitarray_ref }) { next; } print "DISAPEARED hit name: $hit\n"; } push (@previous_hitarray, @{ $oldhitarray_ref }); push (@previous_hitarray, @{ $hitarray_ref }); }

77

Appendix A. Solutions

Solution A.26. Build a Bio::SimpleAlign object

Exercise 4.21
use use use use use Bio::SeqIO; Bio::Tools::Run::StandAloneBlast; Bio::AlignIO; Bio::SimpleAlign; Bio::LocatableSeq; => $ARGV[0],

my $query1_in = Bio::SeqIO->newFh ( -file -format => fasta ); my $query1 = <$query1_in>; my $query2_in = Bio::SeqIO->newFh ( -file -format => fasta ); my $query2 = <$query2_in>;

=> $ARGV[1],

my $out = Bio::AlignIO->newFh(-format => clustalw ); $factory = Bio::Tools::Run::StandAloneBlast->new( program => blastp ); $report = $factory->bl2seq($query1, $query2); while(my $hsp = $report->next_feature) { my $aln = Bio::SimpleAlign->new(); my $querySeq = Bio::LocatableSeq->new( -seq => $hsp->qs, -id => $query1->display_id, -start => 1, -end => $query1->length ); my $sbjctSeq = Bio::LocatableSeq->new( -seq => $hsp->ss, -id => $report->sbjctName, -start => 1, -end => $query2->length ); $aln->add_seq($querySeq); $aln->add_seq($sbjctSeq); print $out $aln; }

A.4. Databases

78

Appendix A. Solutions

Solution A.27.
Exercise 5.1

(a) database construction:

#! /local/bin/perl -w # (a): Genscan, build a database containing one entry per CDS use strict; use Bio::Tools::Genscan; use Bio::SeqIO; my $genscan_file = shift @ARGV; # genscan report my $genscan = Bio::Tools::Genscan->new(-file => $genscan_file); # database to construct my $banque = Bio::SeqIO->newFh ( -fh -format => embl); # date of construction my $date = gmtime; while(my $gene = $genscan->next_prediction()) { # create entry object as database sequence my $entry = Bio::Seq::RichSeq->new(); $entry->add_date($date); $entry->molecule(mRNA); $entry->division(PLN); # extract id of genscan prediction my $cds = $gene->predicted_cds; $cds->id() =~ /\|(.+)\|/; my $id = $1 . " "; # bug $cds->id($id); # add cds as sequence to the entry $entry->primary_seq($cds); # annotation # add all detected exons as SeqFeature to the entry foreach my $exon ($gene->exons()) { $entry->add_SeqFeature($exon); }

=> \*STDOUT,

79

Appendix A. Solutions

# add all detected UTRs as SeqFeature to the entry foreach my $utr ($gene->utrs()) { $entry->add_SeqFeature($utr); } # print the entry to the database flatfile in embl format print $banque $entry; }

You can use this script like this:

perl solution/build_db.pl data/genscan-arab.out > mydb.embl

(b) make index:

#! /usr/bin/perl -w # (b): Genscan, create the index use strict; use Bio::Index::EMBL; my $Index_File_Name = shift; my $inx = Bio::Index::EMBL->new( -filename => $Index_File_Name, -write_flag => 1); $inx->make_index(@ARGV);

You can use this script like this:

perl solution/make_index.pl mydb.idx mydb.embl

80

Appendix A. Solutions

(c) use index:

#! /usr/bin/perl -w # (c): Genscan, retrieve some files use strict; use Bio::Index::EMBL; use Bio::SeqIO; my $out = Bio::SeqIO->newFh ( -fh => \*STDOUT, -format => fasta); my $Index_File_Name = shift; my $inx = Bio::Index::EMBL->new($Index_File_Name); foreach my $id (@ARGV) { my $seq = $inx->fetch($id); if (defined $seq) { print $out $seq; } else { print STDERR "no entry with id ", $id, " in this banque\n."; } }

You can use this script like this:

perl solution/use_index.pl mydb.idx GENSCAN_predicted_CDS_3

since entries names follows the pattern: GENSCAN_predicted_CDS_xxx.

Solution A.28. Parse a Genscan report and build a database entry with the genomic sequence
Exercise 5.2
#! /local/bin/perl -w # Genscan, build a database # - 1 entry per gene with genomic DNA # + 1 feature for the whole CDS # + 1 feature per exon use strict; use Bio::Tools::Genscan; use Bio::SeqIO;

81

Appendix A. Solutions

my $genscan_file = shift @ARGV; my $seqin = Bio::SeqIO->new ( -fh => \*STDIN, -format => fasta); # sequence submitted to genscan my $seq = $seqin->next_seq(); # genscan report my $genscan = Bio::Tools::Genscan->new(-file => $genscan_file); # database to construct my $banque = Bio::SeqIO->newFh ( -fh -format => embl);

=> \*STDOUT,

# +- delta of the subsequence for each entry in database my $delta = 100; # date of construction my $date = gmtime; # print $banque $seq; while(my $gene = $genscan->next_prediction()) { # create entry object as database sequence my $entry = Bio::Seq::RichSeq->new(); $entry->add_date($date); $entry->molecule(DNA); $entry->division(PLN); # extract id of genscan prediction my $cds = $gene->predicted_cds; my @ids = split(/\|/, $cds->id()); # annotation my $start = 0; my $end = 0; my $loc; my $first = 1; my $newloc = new Bio::Location::Split(); my $strand = 1; my $nocds = 0; foreach my $exon ($gene->exons()) { $loc = $exon->location(); if ($first) { $first=0; $start=$loc->start-$delta; } if ($start < 0) { $start = 0; } $loc->start($loc->start-$start); $loc->end ($loc->end-$start);

82

Appendix A. Solutions

$exon->add_tag_value(source, genscan); $exon->primary_tag =~ m/(.*)Exon/; $exon->primary_tag(Exon); $exon->add_tag_value(lc($1), 1); # add all detected exons as SeqFeature to the entry $entry->add_SeqFeature($exon); # construct location for CDS if ($strand == -1 && $loc->strand == 1) { $newloc->warn("exons are coded on different strands"); $nocds = 1; } if ($loc->strand == -1) { $loc->strand(1); $newloc->strand(-1); } $newloc->add_sub_Location($loc); } $end = $loc->end+$delta; if ($end > $seq->length()) { $end = $seq->length(); } if (! $nocds) { # contruct a SeqFeature CDS for the entry ( join all exons) my $newcds = Bio::SeqFeature::Generic->new ( -primary => CDS, -location => $newloc); # add the translation to the CDS Feature $newcds->add_tag_value (translation, $gene->predicted_protein()->seq()); # add the CDS as Feature to the entry $entry->add_SeqFeature($newcds); } # construct the sequence for the entry # subsequence of the sequence submitted to genscan +- delta at each side my $seqstr; if ( $start > $end ) { $seqstr = $seq->subseq($end, $start); } else { $seqstr = $seq->subseq($start, $end); } my $newseq = Bio::PrimarySeq->new ( -seq -id => $ids[1] . " ", -moltype => dna); $entry->primary_seq($newseq); => $seqstr,

# print the entry to the database flatfile in embl format print $banque $entry; }

83

Appendix A. Solutions

Solution A.29. Build a small bioperl module (for the golden program)
Exercise 5.3

# Build a small bioperl module package LocalDBAccess; use strict; use vars qw(@ISA

%AVAIL_LOCALDB);

use Bio::DB::RandomAccessI; use Bio::SeqIO; #use Bio::Seq; BEGIN { %AVAIL_LOCALDB = ( sprot => swiss, sptrnrdb => swiss, gb => genbank, embl => embl ); }

@ISA = qw(Bio::DB::RandomAccessI); sub get_Seq_by_id{ my ($self, @args) = @_; return $self->get_Seq( @args, -i); } sub get_Seq_by_acc{ my ($self, @args) = @_; return $self->get_Seq(@args, -a); } sub get_Seq { my ($self, $entry, $access) = @_;

84

Appendix A. Solutions

my ($db, $id) = split(:, $entry); if ($access !~ /^-[ia]$/) { $access = ""; } if (!$id) { $self->throw ("no database specified"); } if (!_has_local_DB($db)) { $self->throw ("$db -- no such database available"); } my $in = Bio::SeqIO->new( -file => "golden $access $entry 2> /dev/null |", -format => _get_seq_format($db)); my $seq = $in->next_seq(); # # # # $in->DESTROY(); if ($? >> 8) { $self->throw ("$id -- no such entry in database $db"); } $self->throw ("$id -- no such entry in database $db") if (! defined $seq); return $seq; }

sub _has_local_DB { my ($db) = @_; if (exists $AVAIL_LOCALDB{$db}) { return 1; } return 0; } sub _get_seq_format { my ($db) = @_; return $AVAIL_LOCALDB{$db}; }

85

Appendix A. Solutions

86