1 #summary Tutorial for multiple sequence alignments and phylogenetic methods in BioRuby -- under development!
9 Tutorial for multiple sequence alignments and phylogenetic methods in [http://bioruby.open-bio.org/ BioRuby].
11 Eventually, this is expected to be placed on the official !BioRuby page.
13 Author: [http://www.cmzmasek.net/ Christian M Zmasek], Sanford-Burnham Medical Research Institute
16 Copyright (C) 2011 Christian M Zmasek. All rights reserved.
19 = Multiple Sequence Alignment =
22 == Multiple Sequence Alignment Input and Output ==
24 === Reading in a Multiple Sequence Alignment from a File ===
26 ==== ClustalW Format ====
28 The following example shows how to read in a *ClustalW*-formatted multiple sequence alignment.
34 # Reads in a ClustalW-formatted multiple sequence alignment
35 # from a file named "infile_clustalw.aln" and stores it in 'report'.
36 report = Bio::ClustalW::Report.new(File.read('infile_clustalw.aln'))
38 # Accesses the actual alignment.
39 msa = report.alignment
41 # Goes through all sequences in 'msa' and prints the
42 # actual molecular sequence.
48 ==== FASTA Format ====
50 The following example shows how to read in a *FASTA*-formatted multiple sequence file. (_This seems a little clumsy, I wonder if there is a more direct way, avoiding the creation of an array.)
55 # Reads in a FASTA-formatted multiple sequence alignment (which does
56 # not have to be aligned, though) and stores its sequences in
59 fasta_seqs = Bio::Alignment::MultiFastaFormat.new(File.open('infile.fasta').read)
60 fasta_seqs.entries.each do |seq|
64 # Creates a multiple sequence alignment (possibly unaligned) named
65 # 'seqs' from array 'seq_ary'.
66 seqs = Bio::Alignment.new(seq_ary)
68 # Prints each sequence to the console.
69 seqs.each { |seq| puts seq.to_s }
71 # Writes multiple sequence alignment (possibly unaligned) 'seqs'
72 # to a file in PHYLIP format.
73 File.open('out1.phylip', 'w') do |f|
74 f.write(seqs.output(:phylip))
78 Relevant API documentation:
80 * [http://bioruby.open-bio.org/rdoc/classes/Bio/ClustalW/Report.html Bio::ClustalW::Report]
81 * [http://bioruby.open-bio.org/rdoc/classes/Bio/Alignment.html Bio::Alignment]
82 * [http://bioruby.open-bio.org/rdoc/classes/Bio/Sequence.html Bio::Sequence]
84 === Writing a Multiple Sequence Alignment to a File ===
87 The following example shows how to write a multiple sequence alignment in *FASTA*-format. It first creates a file named "outfile.fasta" for writing ('w') and then writes the multiple sequence alignment referred to by variable 'msa' to it in FASTA-format (':fasta').
93 # Creates a new file named "outfile.fasta" and writes
94 # multiple sequence alignment 'msa' to it in fasta format.
95 File.open('outfile.fasta', 'w') do |f|
96 f.write(msa.output(:fasta))
100 ==== Setting the Output Format ====
102 The following symbols determine the output format:
104 * `:clustal` for ClustalW
106 * `:phylip` for PHYLIP interleaved (will truncate sequence names to no more than 10 characters)
107 * `:phylipnon` for PHYLIP non-interleaved (will truncate sequence names to no more than 10 characters)
109 * `:molphy` for Molphy
112 For example, the following writes in PHYLIP's non-interleaved format:
115 f.write(align.output(:phylipnon))
119 === Formatting of Individual Sequences ===
121 !BioRuby can format molecular sequences in a variety of formats.
122 Individual sequences can be formatted to (e.g.) Genbank format as shown in the following examples.
124 For Sequence objects:
126 seq.to_seq.output(:genbank)
129 For Bio::!FlatFile entries:
131 entry.to_biosequence.output(:genbank)
134 The following symbols determine the output format:
135 * `:genbank` for Genbank
138 * `:fasta_ncbi` for NCBI-type FASTA
139 * `:raw` for raw sequence
140 * `:fastq` for FASTQ (includes quality scores)
141 * `:fastq_sanger` for Sanger-type FASTQ
142 * `:fastq_solexa` for Solexa-type FASTQ
143 * `:fastq_illumina` for Illumina-type FASTQ
145 == Calculating Multiple Sequence Alignments ==
147 !BioRuby can be used to execute a variety of multiple sequence alignment
148 programs (such as [http://mafft.cbrc.jp/alignment/software/ MAFFT], [http://probcons.stanford.edu/ Probcons], [http://www.clustal.org/ ClustalW], [http://www.drive5.com/muscle/ Muscle], and [http://www.tcoffee.org/Projects_home_page/t_coffee_home_page.html T-Coffee]).
149 In the following, examples for using the MAFFT and Muscle are shown.
154 The following example uses the MAFFT program to align four sequences
155 and then prints the result to the screen.
156 Please note that if the path to the MAFFT executable is properly set `mafft=Bio::MAFFT.new(options)` can be used instead of explicitly indicating the path as in the example.
162 # 'seqs' is either an array of sequences or a multiple sequence
163 # alignment. In general this is read in from a file as described in ?.
164 # For the purpose of this tutorial, it is generated in code.
165 seqs = ['MFQIPEFEPSEQEDSSSAER',
166 'MGTPKQPSLAPAHALGLRKS',
171 # Calculates the alignment using the MAFFT program on the local
172 # machine with options '--maxiterate 1000 --localpair'
173 # and stores the result in 'report'.
174 options = ['--maxiterate', '1000', '--localpair']
175 mafft = Bio::MAFFT.new('path/to/mafft', options)
176 report = mafft.query_align(seqs)
178 # Accesses the actual alignment.
179 align = report.alignment
181 # Prints each sequence to the console.
182 align.each { |s| puts s.to_s }
188 * Katoh, Toh (2008) "Recent developments in the MAFFT multiple sequence alignment program" Briefings in Bioinformatics 9:286-298
190 * Katoh, Toh 2010 (2010) "Parallelization of the MAFFT multiple sequence alignment program" Bioinformatics 26:1899-1900
200 # 'seqs' is either an array of sequences or a multiple sequence
201 # alignment. In general this is read in from a file as described in ?.
202 # For the purpose of this tutorial, it is generated in code.
203 seqs = ['MFQIPEFEPSEQEDSSSAER',
204 'MGTPKQPSLAPAHALGLRKS',
208 # Calculates the alignment using the Muscle program on the local
209 # machine with options '-quiet -maxiters 64'
210 # and stores the result in 'report'.
211 options = ['-quiet', '-maxiters', '64']
212 muscle = Bio::Muscle.new('path/to/muscle', options)
213 report = muscle.query_align(seqs)
215 # Accesses the actual alignment.
216 align = report.alignment
218 # Prints each sequence to the console.
219 align.each { |s| puts s.to_s }
225 * Edgar, R.C. (2004) "MUSCLE: multiple sequence alignment with high accuracy and high throughput" Nucleic Acids Res 32(5):1792-1797
227 === Other Programs ===
229 _need more detail here..._
231 [http://probcons.stanford.edu/ Probcons], [http://www.clustal.org/ ClustalW], and [http://www.tcoffee.org/Projects_home_page/t_coffee_home_page.html T-Coffee] can be used in the same manner as the programs above.
235 == Manipulating Multiple Sequence Alignments ==
237 Oftentimes, multiple sequence to be used for phylogenetic inference are 'cleaned up' in some manner. For instance, some researchers prefer to delete columns with more than 50% gaps. The following code is an example of how to do that in !BioRuby.
251 = Phylogenetic Trees =
254 == Phylogenetic Tree Input and Output ==
257 === Reading in of Phylogenetic Trees ===
261 ====Newick or New Hampshire Format====
271 ====phyloXML Format====
273 Partially copied from [https://www.nescent.org/wg_phyloinformatics/BioRuby_PhyloXML_HowTo_documentation Diana Jaunzeikare's documentation].
275 In addition to !BioRuby, a libxml Ruby binding is also required. This can be installed with the following command:
278 % gem install -r libxml-ruby
281 This example reads file "example.xml" and stores its [http://www.phyloxml.org/ phyloXML]-formatted trees in variable 'trees'.
287 # This creates new phyloXML parser.
288 trees = Bio::PhyloXML::Parser.new('example.xml')
290 # This prints the names of all trees in the file.
295 # If there are several trees in the file, you can access the one you wish via index.
311 === Writing of Phylogenetic Trees ===
313 ====Newick or New Hampshire Format====
323 ====phyloXML Format====
325 Partially copied from [https://www.nescent.org/wg_phyloinformatics/BioRuby_PhyloXML_HowTo_documentation Diana Jaunzeikare's documentation].
327 In addition to !BioRuby, a libxml Ruby binding is also required. This can be installed with the following command:
330 % gem install -r libxml-ruby
333 This example writes trees 'tree1' and 'tree2' to file "tree.xml" in [http://www.phyloxml.org/ phyloXML] format.
339 # this creates new phyloXML writer.
340 writer = Bio::PhyloXML::Writer.new('tree.xml')
342 # Writes tree to the file "tree.xml".
345 # Adds another tree to the file.
363 = Phylogenetic Inference =
365 _Currently !BioRuby does not contain wrappers for phylogenetic inference programs, thus I am progress of writing a RAxML wrapper followed by a wrapper for FastME..._
367 == Optimality Criteria Based on Character Data ==
369 Character data based methods work directly on molecular sequences and thus do not require the calculation of pairwise distances but tend to be time consuming and sensitive to errors in the multiple sequence alignment.
371 === Maximum Likelihood ===
394 === Maximum Parsimony ===
396 Currently no direct support in !BioRuby.
399 === Bayesian Inference ===
401 Currently no direct support in !BioRuby.
404 == Pairwise Distance Based Methods ==
406 === Pairwise Sequence Distance Estimation ===
417 === Optimality Criteria Based on Pairwise Distances ===
420 ==== Minimal Evolution: FastME ====
430 === Algorithmic Methods Based on Pairwise Distances ===
432 ==== Neighbor Joining and Related Methods ====
448 == Support Calculation? ==
450 === Bootstrap Resampling? ===
455 = Analyzing Phylogenetic Trees =
460 == Gene Duplication Inference ==
462 _need to further test and then import GSoC 'SDI' work..._
470 = Putting It All Together =
472 Example of a small "pipeline"-type program running a mininal phyogenetic analysis: starting with a set of sequences and ending with a phylogenetic tree.