CRESS-GLUE

Using CRESS-GLUE - an example-driven tutorial

This tutorial focuses on beak and feather disease virus and comprises the following steps:

1. Downloading selected virus sequences from GenBank.
2. Linking virus sequences with other data and extracting isolate data from GenBank files.
3. Filtering sequence data to obtain a valid analysis set.
4. Defining virus reference sequences and associating them with annotations.
5. Standardising sequence orientation and constructing in-frame alignments.
6. Mapping feature coverage of sequence members within an alignment.
7. Examining genetic variation in protein-coding genes.
8. Reconstructing phylogenetic trees: bootstrapped, rooted, and annotated.
9. Using GLUE projects as a basis for collaborative research and resource development.

If you haven't decided yet whether or not you want to use GLUE, you may find it useful to browse through this tutorial - it should provide some insight into what GLUE offers and the kind of things that can be done with GLUE projects.

If you've already committed to using GLUE, you may find it useful to also explore the GLUE example project when following this tutorial.

Background - beak and feather disease virus, a pathogenic circovirus

Left to right: the African Grey (Psittacus erithacus); blue-and-yellow macaw (Ara ararauna); scarlet macaw (Ara macao); budgerigar (Melopsittacus undulatus).

Psittacine birds (parrots and cockatoos) are among the most beautiful and remarkable creatures on the planet.

Unfortunately, many of these wonderful species are presently endangered, and conservation efforts have been hampered by the emergence of psittacine beak and feather disease (PBFD). PBFD is caused by a circovirus (genus Circovirus) called beak and feather disease virus (BFDV).

Circovirus genomes encode two major proteins: replication-associated protein (Rep) and capsid (Cap), responsible for genome replication and particle formation respectively. Transcription is bidirectional with the rep gene being encoded in the forward sense, and the cap gene being encoded in the complementary sense.

Genomic approaches have great potential to help control the spread of BFDV. In this tutorial we will use the CRESS-GLUE project to (i) investigate BFDV distribution diversity and evolution, and (ii) establish a sequence-based resource for BFDV.

1. Downloading all available BFDV sequences from GenBank

To download all BFDV entries in NCBI GenBank, we will use a customised version of GLUE's module 'ncbiImporter' module. Our project-specific configuration of the module can be viewed here. Viewing the file, you can probably see that it is configured to download sequences based on a query phrase:

"Beak and feather disease virus"[Organism] AND 200:2000[SLEN]

This 'eSearchTerm' is a standard NCBI entrez text query that specifies all GenBank entries labelled "beak and feather disease" in the 'Organism' field and between 200-2000 nucleotides (nt) in length.

To use the module, first initiate GLUE on the command line as follows:

  MyComputer:CRESS-GLUE rob$ gluetools.sh 
  GLUE Version 1.1.103
  Copyright (C) 2015-2020 The University of Glasgow
  This program comes with ABSOLUTELY NO WARRANTY. This is free software, and you
  are welcome to redistribute it under certain conditions. For details see
  GNU Affero General Public License v3: http://www.gnu.org/licenses/

  Mode path: /
  GLUE>

Next, navigate to the 'cress' project as shown here:

  GLUE> project cress 
  OK
  Mode path: /project/cress/

Now create the module using it's configuration file (which is contained in the CRESS-GLUE project) as shown here:

  Mode path: /project/cress/
  GLUE> create module -f modules/build/genus/circoviridae/circo/circoviridaeNcbiImporterBfdv.xml
  OK
  (1 Module created) 

To run the module, execute the following command in the GLUE shell:

  Mode path: /project/cress/
  GLUE> module circoviridaeNcbiImporterBfdv import

When I ran this command on the 1st December 2020, I obtained the following output:

  ncbiImporterSummaryResult
  totalMatching: 1250
  present: 0
  surplus: 0
  missing: 1250
  deleted: 0
  downloaded: 1250

Now export the sequences to a 'source' directory as follows.

   Mode path: /project/cress
   GLUE> export source ncbi-curated-circo-bfdv -p sources/

GLUE has been designed to provide as much flexibility as possible with respect to data-formatting. To export FASTA-formatted sequences from GLUE, use a 'fastaExporter' module.

As with many of GLUE's core module types, a version with a default configuration can be created on the GLUE command line using the '-t' (template) option:

   Mode path: /project/cress
   GLUE> create module -t fastaExporter 
   OK
   (1 Module created)

Now lets use this module to export the sequences in the 'ncbi-curated-circo-bfdv' source. Heres my first attempt:

  Mode path: /project/cress
  GLUE> module fastaExporter export -w "source.name = 'ncbi-curated-circo-bfdv'"
  Error: Usage error
  Usage: export (-w <whereClause> | -a) [-o <offset> -b <batchSize>] [-y <lineFeedStyle>]
                [-r] [-t] (-p | -f <fileName>)

  Options:
	-y <lineFeedStyle>, --lineFeedStyle <lineFeedStyle>  LF or CRLF
	-f <fileName>, --fileName <fileName>                 FASTA file
	-w <whereClause>, --whereClause <whereClause>        Qualify exported sequences
	-o <offset>, --offset <offset>                       Paged query offset
	-b <batchSize>, --batchSize <atchSize>               Paged query batch size
	-a, --allSequences                                   Export all project sequences
	-r, --suppressReverseComplement                      Suppress reverse complement
	-t, --suppressRotation                               Suppress rotation
	-p, --preview

Oops - something went wrong! - the module needs another argument.

Note that the error message prompts with the correct usage of the module, and an explanation of the options. In this case, I forgot to tell GLUE what to do with the sequences. They can be previewed in the console, or exported to a file. Run the module as follows to export the sequences to a file named 'bfdv.fna':

  Mode path: /project/cress
  GLUE> module fastaExporter export -w "source.name = 'ncbi-curated-circo-bfdv'" -f bfdv.fna
  OK

You can check how to use any of the commands associated with a GLUEs module by using the 'help' option.

  Mode path: /project/cress
  GLUE> module fastaExporter help export

2. Importing BFDV sequences and metadata from GenBank files.

Now that we've downloaded a set of BFDV sequences, we can incorporate them into the CRESS-GLUE project. Load the sequences as follows:

  Mode path: /project/cress
  GLUE> import source sources/ncbi-curated-circo-bfdv/

Once loaded, the sequences can be viewed in the project using the 'list sequence' command:

  Mode path: /project/cress
  GLUE> list sequence sequenceID source.name family genus -w "source.name = 'ncbi-curated-circo-bfdv'"
  +============+=========================+=========+=========+
  | sequenceID |       source.name       | family  |  genus  |
  +============+=========================+=========+=========+
  | AB182568   | ncbi-curated-circo-bfdv |    -    |    -    |
  | AB182569   | ncbi-curated-circo-bfdv |    -    |    -    |
  | AB182570   | ncbi-curated-circo-bfdv |    -    |    -    |
  | AB182571   | ncbi-curated-circo-bfdv |    -    |    -    |
  | AB182572   | ncbi-curated-circo-bfdv |    -    |    -    |

The database fields associated with these side data are defined in CRESS-GLUE's project-specific schema extensions.

All BFDV sequences have identical taxonomic information down to species level. An efficient way to input the taxonomic data for each sequence is to use this script. The script comprises a single loop, and 'wraps' two GLUE commands. The first command lists all the sequences in our BFDV-specific source directory. Next we step through this list of sequences, and use GLUE's 'set' command to setting the appropriate values in taxonomic data fields in the corresponding table row.

Run the script as follows:

  Mode path: /project/cress
  GLUE> run script glue/build/genus/circo/exampleLoadSequenceDataBfdv.js

Inspect data in the sequence table via the command line:

  Mode path: /project/cress
  GLUE> list sequence sequenceID name full_name family genus clade
  +============+===============+================+==================+==============+
  | sequenceID |     name      |     family     |      genus       |    clade     |
  +============+===============+================+==================+==============+
  | AB182568   | BFDV          | Circoviridae   | Circovirus       | Avian-1      |
  | AB182569   | BFDV          | Circoviridae   | Circovirus       | Avian-1      |
  | AB182570   | BFDV          | Circoviridae   | Circovirus       | Avian-1      |
  | AB182571   | BFDV          | Circoviridae   | Circovirus       | Avian-1      |
  | AB182572   | BFDV          | Circoviridae   | Circovirus       | Avian-1      |

Next, we will extract BFDV isolate information from GenBank files. First, we need to link each new BFDV sequence to the 'isolate table'. To do this, we'll use this script.

 Mode path: /project/cress
 GLUE> run script glue/build/genus/circo/exampleLoadIsolateDataBfdv.js

GLUE's genbankPopulator module can be used to extract sequence and isolate information from GenBank files. We can use this module to search for for host species, time and place of isolation, and isolate data in the "notes" section of the GenBank XML.

 Mode path: /project/cress
 GLUE> module genbankPopulator populate

Inspect data in the isolate table via the command line. In the example below a 'where clause' is used to limit the query to Japanese isolates only:

 Mode path: /project/cress
 GLUE> list custom-table-row isolate id isolate_id country -w "country = 'Japan'"
 +==========+============+=========+
 |    id    |  isolate   | country |
 +==========+============+=========+
 | AB277726 | AGP-SA1P-1 | Japan   |
 | AB277727 | AGP-SA1P-2 | Japan   |
 | AB277728 | AGP-SA1P-3 | Japan   |
 | AB277729 | AGP-SA1P-4 | Japan   |
 | AB277730 | AGP-SA1P-5 | Japan   |

Executing the query above without a where clause will show that not every sequence has complete information for each field in the isolate table. This is because typically, some GenBank entries for any given virus lack this information.

3. Filtering data to obtain a valid analysis data set.

Investigating the data in the sequence table reveals that some of the BFDV sequences are not associated with isolate data. This means that our genbankXmlPopulator module didn't find the information in the XML file for that sequence.

Inconsistent annotation of virus entries in GenBank is fairly commonplace. The GenBank database contains sequence data from diverse sources (i.e. not just from viruses) Consequently, the way in which isolate-related information is recorded in GenBank files is not highly standardised, and ofte data are missing.

It's often possible to fill in missing values via reference to an associated manuscript. In this section I'll show how to use GLUE to coordinate this process.

First, let's export the existing isolate table data for sequence in our source. This is a three step process:

  Mode path: /project/cress
  GLUE> console set cmd-output-file-format tab 
  OK
  
  Mode path: /project/cress
  GLUE> console set next-cmd-output-file bfdv-data.tsv
  OK  
  Mode path: /project/cress
  GLUE> list sequence sequenceID isolate.isolate_id isolate.country isolate.host_sci_name isolate.collection_year -w "source.name = 'ncbi-curated-circo-bfdv'"
  OK

Take a look at the first few rows of data:

  +============+========================+========================+=========================+
  | sequenceID |   isolate.isolate_id   |    isolate.country     |  isolate.host_sci_name  |
  +============+========================+========================+=========================+
  | AB182568   |YCC-IN1P                |-                       |-                        |
  | AB182569   |YCC-IN2P                |-                       |-                        |
  | AB182570   |RRP-JP11P               |-                       |-                        |
  | AB182571   |RRP-JP12P               |-                       |-                        |
  | AB182572   |BTP-SA11P               |-                       |-                        |
  | AB182573   |RL-JP1P                 |-                       |-                        |
  | AB277726   |AGP-SA1P-1              |Japan                   |Psittacus erithacus      |

Looking at the corresponding GenBank entry on the NCBI website - e.g. this one for sequence AB182568 - confirms that the information is not contained within the GenBank entry.

However, the manuscript associated with these sequences includes a table (Table 1) that contains the missing information

The next few steps will demonstrate how we can improve on the annotations in GenBank by supplementing missing information. First let's take a look at the data file we exported (click here)

We will use a slight altered version of this file to upload the refined data - (click here). It is basically the same file but for each of the data columns (other than the sequence ID column) we have created a duplicate column to record an update - e.g. for 'country' we have 'country_updated'.

The updated columns are for recording any changes in the data, if no change is made, the value will always be 'unchanged'. Note that in the updated file the first 'country' and 'host_sci_name' fields have been updated in the rows in corresponding to accession numbers AB182568-AB182572.

Once the edits have been made, use this this script to apply the changes in CRESS-GLUE. The updated information can now be viewed, as follows:

  Mode path: /project/cress
  GLUE> list sequence sequenceID isolate.isolate_id isolate.country isolate.host_sci_name -w "source.name = 'ncbi-curated-circo-bfdv'"
  +============+=============================+=============================+=============================+
  | sequenceID |     isolate.isolate_id      |       isolate.country       |    isolate.host_sci_name    |
  +============+=============================+=============================+=============================+
  | AB182568   |YCC-IN1P                     |Indonesia                    |Cacatua sulphurea            |
  | AB182569   |YCC-IN2P                     |Indonesia                    |Cacatua sulphurea            |
  | AB182570   |RRP-JP11P                    |Japan                        |Psittacula krameri           |
  | AB182571   |RRP-JP12P                    |Japan                        |Psittacula krameri           |
  | AB182572   |BTP-SA11P                    |South Africa                 |Aratinga pertinax            |
  | AB182573   |RL-JP1P                      |Japan                        |Eos bornea                   |
  | AB277726   |AGP-SA1P-1                   |Japan                        |Psittacus erithacus erithacus|

4. Refining the BFDV reference sequence.

The 'Circovirus' extension of CRESS-GLUE includes reference sequences for most (if not all) species, including BFDV. The reference sequence declarations can be found in this file.

However, in a project focussed specifically on BFDV, we may want to annotate the BFDV more richly than it is presently annotated within the 'circovirus' component of CRESS-GLUE. In fact, it makes sense to incorporate everything that is known about the BFDV genome.

Let's examine how we can modify the existing BFDV reference sequence. First, let's have a look at the reference sequences in the cress project. On way to do this is to use the tab complete function of the console to reveal the full list of reference sequences, by first typing 'reference (or just the first few letters of this word), and then pressing the 'tab' key, as shown here:

 Mode path: /project/cress
 GLUE> reference REF_
 REF_Begomo_EACMKV              REF_CRESS1_MASTER_AHEaCV-13    REF_CRESS2_MASTER_Po-41        
 REF_Circo_BFDV                 REF_Circo_BarbCV               REF_Circo_CaCV                 
 REF_Circo_CfCV                 REF_Circo_CoCV                 REF_Circo_DuCV                 
 REF_Circo_FiCV                 REF_Circo_GuCV                 REF_Circo_MASTER_PCV-1         
 REF_Circo_MiCV                 REF_Circo_PCV-2                REF_Circo_PCV-3                
 REF_Circo_RaCV                 REF_Circo_SgCV                 REF_Circo_StCV                 
 REF_Circo_SvCV                 REF_Circo_SwCV                 REF_Circo_TbCV                 
 REF_Circo_ZfCV                 REF_Cyclo_MASTER_HuACyV-1      REF_Porprismaco_MASTER_CBSV    
 REF_Redondo_MASTER_HLVV

As expected, we see the master references of the core project, plus the references associated with the Circovirus extension. To investigate further, navigate to the BFDV reference sequence and list all the features that are presently associated with this reference, as follows:

  Mode path: /project/cress
  GLUE> reference REF_Circo_BFDV 
  OK
  Mode path: /project/cress/reference/REF_Circo_BFDV
  GLUE> list feature-location 
  +========================+==============+
  | referenceSequence.name | feature.name |
  +========================+==============+
  | REF_Circo_BFDV         | whole_genome |
  | REF_Circo_BFDV         | Rep          |
  | REF_Circo_BFDV         | Cap          |
  +========================+==============+
  FeatureLocations found: 4

The reference has the locations of major genome features mapped - specifically the rep and cap genes. Inspecting the BFDV GenBank entry shows that includes annotation for an additional, hypothetical open reading frame (ORF) called "V2".

Let's modify the existing BFDV reference ssequence to include V2. First, lets examine all of the features defined in the entire CRESS-GLUE project.

  GLUE> project-mode 
  OK
  Mode path: /project/cress
  GLUE> list feature
  +==============+==============+===================================+
  |     name     | parent.name  |            description            |
  +==============+==============+===================================+
  | whole_genome | -            | Whole genome                      |
  | 5flank       | -            | 5 prime genomic flanking sequence |
  | 3flank       | -            | 3 prime genomic flanking sequence |
  | Rep          | whole_genome | Replicase protein                 |
  | ORFV2        | whole_genome | Hypothetical protein ORFV2        |
  | ORFC2        | whole_genome | Hypothetical protein ORFC2        |
  | ORF4         | whole_genome | Hypothetical protein ORF4         |
  | ORF3         | whole_genome | Hypothetical protein ORF3         |
  | nonanuc      | whole_genome | nonanucleotide motif              |
  | intergenic   | whole_genome | intergenic region                 |
  | gp3          | whole_genome | gp3 protein                       |
  | gp2          | whole_genome | gp2 protein                       |
  | Cap          | whole_genome | Capsid protein                    |
  | 5UTR         | whole_genome | 5' UTR                            |
  | 3UTR         | whole_genome | 5' 3TR                            |
  +==============+==============+===================================+
  Features found: 15

First let's create a protein-coding genome feature called V2, as follows:

  Mode path: /project/cress
  GLUE> create feature v2 "v2 gene"
  OK
  (1 Feature created)
  
  GLUE> feature v2 set metatag CODES_AMINO_ACIDS true
  OK
  
  Mode path: /project/cress/
  GLUE> feature v2 set metatag OWN_CODON_NUMBERING true
  OK

Now lets associate this feature with a location in the BFDV genome. The first step is to create the feature location in the BFDV reference sequence:

  Mode path: /project/cress
  GLUE> reference REF_Circo_BFDV add feature-location v2
  OK
  (1 FeatureLocation created)

Now we need to associate the feature location with the appropriate region of the BFDV genome, as follows:

  GLUE> reference REF_Circo_BFDV feature-location v2 add segment 550 1026
  OK
  (1 FeatureSegment created)

The coordinates (550-1026) are derived from the GenBank entry for our BFDV reference sequence.

As always, the same result can be accomplished by running a file that contains the GLUE commands, as follows:

  Mode path: /project/cress
  GLUE> run file glue/build/genus/circo/exampleSetFeatureLocationsBfdv.glue

By adding this feature to the BFDV we have essentially given CRESS-GLUE 'knowledge' of this genome feature - meaning that we will now be able to reference it when we are working with the BFDV sequences in GLUE.

For example, we can export a v2 alignment partition, or use the alignment within GLUE to examing protein-level variation in the v2 gene. First, however, we will need to create an alignment, which is our next step.

Left to right: palm cockatoo (Probosciger aterrimuss); rainbow lorikeet (Trichoglossus moluccanus); budgerigar (Melopsittacus undulatus); kakapo (Strigops habroptilus).

5. Creating an alignment of all BFDV sequences.

Now we have put together the fundamental data items we can begin working on some proper comparative analysis. In the next few steps we will use everything we've created so far to start investigating the genetic diversity of BFDV in relation to it's biology.

Most comparative studies of virus genomes require the creation of a multiple sequence alignment containing the sequences under investigation.

Large set of sequences derived from a single virus species can be efficiently aligned to one another using a pairwise, reference-guided approach. The GLUE framework allows this approach to be used in a codon-aware way, and without data loss.

At this point, I should mention that some BFDV sequences are in the opposite orientation to our chosen BFDV reference. To address this, run this GLUE file which will effectively tell the project which sequences need to be reverse complemented before they are aligned. This will allow us to proceed directly to the alignment step with all sequences in the same orientation. Later in this section, I'll explain how I created the file.

  run file glue/build/genus/circo/exampleApplyReverseComplement.glue

To create an alignment, first, exit the reference sequence to project level (assuming you are still in this mode path), as follows:

   Mode path: /project/cress/reference/REF_Circo_BFDV
   GLUE> project-mode 
   OK
   Mode path: /project/cress
   GLUE>

Before we perform the alignment construction, we need to first create a constrained alignment object and specify which sequences belong in it, as shown here:

  Mode path: /project/cress
  GLUE> create alignment AL_BFDV -r REF_Circo_BFDV
  OK
  
  Mode path: /project/cress
  GLUE> alignment AL_BFDV add member --whereClause "source.name = 'ncbi-curated-circo-bfdv'"
  OK

Now we can construct the alignment. The CRESS-GLUE project contains a circovirus-specific configuration file for GLUE's 'compoundAligner' module.

  compute alignment AL_BFDV circovirusCompoundAligner

Now let's inspect the resulting alignment. One way is to use the 'show statistics' command. This will produce a table as shown below (note that only the first few table rows are shown here):

  GLUE> alignment AL_BFDV show statistics
  +=========================+=====================+============================+=========================+==========+==========+=============+=============+
  |  sequence.source.name   | sequence.sequenceID | referenceNtCoveragePercent | memberNtCoveragePercent | minRefNt | maxRefNt | minMemberNt | maxMemberNt |
  +=========================+=====================+============================+=========================+==========+==========+=============+=============+
  | ncbi-curated-circo-bfdv | AB182568            | 22.277972905168088         | 99.32885906040268       | 1402     | 1845     | 3           | 446         |
  | ncbi-curated-circo-bfdv | AB182569            | 22.277972905168088         | 99.32885906040268       | 1402     | 1845     | 3           | 446         |
  | ncbi-curated-circo-bfdv | AB182570            | 22.37832413447065          | 99.77628635346755       | 1400     | 1845     | 1           | 446         |
  | ncbi-curated-circo-bfdv | AB182571            | 22.37832413447065          | 99.77628635346755       | 1400     | 1845     | 1           | 446         |
  | ncbi-curated-circo-bfdv | AB182572            | 22.37832413447065          | 99.77628635346755       | 1400     | 1845     | 1           | 446         |

Or alternatively, list alignment members, along with their associated data, like this:

  GLUE> alignment AL_BFDV 
  OK
  Mode path: /project/cress/alignment/AL_BFDV
  GLUE> list member sequence.sequenceID sequence.name sequence.isolate.isolate_id
  +=====================+===============+====================================+
  | sequence.sequenceID | sequence.name |      sequence.isolate.isolate_id   |
  +=====================+===============+====================================+
  | AB182568            | BFDV          | YCC-IN1P                           |
  | AB182569            | BFDV          | YCC-IN2P                           |
  | AB182570            | BFDV          | RRP-JP11P                          |
  | AB182571            | BFDV          | RRP-JP12P                          |
  | AB182572            | BFDV          | BTP-SA11P                          |

(only the first few results are shown here)

So, how did I figure out which sequences needed to be reverse-complemented? There are lots of ways to do this using GLUE. In this case I used a fairly unsophisticated, two-step approach - but one that seemed to work quite well.

First, I naively aligned all BFDV sequences without reverse complementing any of them, then I inspected the alignment in the console<.

Since we had excluded sequences shorter than 200bp when we downloaded from BFDV sequences GenBank (see above), it was a reasonable guess that any sequences for which the only aligned segments were drastically shorter than this length were in reverse orientation relative to the reference.

Additionally, since (i) the orientation of viral sequences is usually consistent within a study, and (ii) sequences from the same study are usually numbered sequentially (e.g. AB182568-AB182573), it wasn't too demanding to create a list of BFDV sequences suspected to require reverse complementation.

The alignment can be exported using a fastaAlignmentExporter module, as shown here:

  Mode path: /project/cress
  GLUE> create module -t fastaAlignmentExporter 
  OK
  (1 Module created)
  Mode path: /project/cress
  GLUE> module fastaAlignmentExporter 
  OK
  Mode path: /project/cress/module/fastaAlignmentExporter
  GLUE> export AL_BFDV -r REF_Circo_BFDV -f v2 -a -e -o v2.aln.fna
  OK

The entire alignment can be exported, or the reference sequence can be used to export specific alignment sections. In the example shown above I've exported the alignment section corresponding to the v2 gene to a file called 'v2.aln.fna'. The '-e' option tells GLUE to exclude any sequences that have no coverage in the v2 gene.

6. Mapping feature coverage in BFDV sequences.

Now that we have created an alignment of BFDV sequences that is constrained to the BFDV reference sequence, we can use this alignment in combination with GLUE's 'featurePresenceRecorder' module to examine feature coverage within individual BFDV sequences.

  Mode path: /project/cress
  GLUE> create module -f modules/build/genus/circoviridae/circo/circoFeaturePresenceRecorder.xml 
  OK
  (1 Module created)
  Mode path: /project/cress
  GLUE> module circoFeaturePresenceRecorder
  OK
  Mode path: /project/cress/module/circoFeaturePresenceRecorder
  GLUE> record feature-presence AL_BFDV -r -f whole_genome -d

This can also be accomplished by running a file containing these commands, as follows:

  Mode path: /project/cress
  GLUE> run file glue/build/genus/circo/exampleRecordFeaturePresenceBfdv.glue 
  OK

7. Examining genetic variation.

To examine genetic variation within a protein coding genome features of the BFDV genome, we can use our constrained alignment of BFDV sequences, and GLUE's 'amino acid frequency' command. The command produces a table report of amino acid variation within coding features, as in the following example:

  Mode path: /project/cress
  GLUE> alignment AL_BFDV amino-acid frequency -r REF_Circo_BFDV -f Rep -l 100 105
  +=========+=======+===========+============+=====================+
  | feature | codon | aminoAcid | numMembers |     pctMembers      |
  +=========+=======+===========+============+=====================+
  | Rep     | 100   | C         | 940        | 99.89373007438894   |
  | Rep     | 100   | W         | 1          | 0.10626992561105207 |
  | Rep     | 101   | A         | 1          | 0.10649627263045794 |
  | Rep     | 101   | I         | 1          | 0.10649627263045794 |
  | Rep     | 101   | R         | 1          | 0.10649627263045794 |
  | Rep     | 101   | S         | 935        | 99.57401490947817   |
  | Rep     | 101   | T         | 1          | 0.10649627263045794 |
  | Rep     | 102   | K         | 940        | 99.78768577494692   |
  | Rep     | 102   | R         | 2          | 0.21231422505307856 |
  | Rep     | 103   | D         | 15         | 1.5940488841657812  |
  | Rep     | 103   | E         | 925        | 98.29968119022317   |
  | Rep     | 103   | R         | 1          | 0.10626992561105207 |
  | Rep     | 104   | G         | 944        | 100.0               |
  | Rep     | 105   | D         | 941        | 99.6822033898305    |
  | Rep     | 105   | E         | 1          | 0.1059322033898305  |
  | Rep     | 105   | N         | 1          | 0.1059322033898305  |
  | Rep     | 105   | R         | 1          | 0.1059322033898305  |
  +=========+=======+===========+============+=====================+

8. Reconstructing phylogenetic trees.

In the final step in this tutorial, we will use the data we have collated in this tutorial to reconstruct bootstrapped, midpoint-rooted, and annotated trees for each of the two main genes in the BFDV genome (Cap and Rep).

We will use 'RAxML' to construct trees, via GLUE's 'raxmlPhylogenyGenerator' module. The entire sequence of commands for creating both trees can be found in this file, which is in the CRESS-GLUE repo. Run this file as follows:

  Mode path: /project/cress
  GLUE> run file glue/build/genus/circo/exampleCreateBfdvPhylogenies.glue

To midpoint-root the trees, we will use GLUE's module. To apply the module to the two trees we've constructed, run this file as follows:

  Mode path: /project/cress
  GLUE> run file glue/build/genus/circo/exampleRerootPhylogeny.glue

Export annotations for the trees by running this file, as follows:

  Mode path: /project/cress
  GLUE> run file glue/build/genus/circo/exampleExportAnnotations.glue

The finished trees with annotations can be found here. One way to view these trees is by using Andrew Rambaut's FigTree program.

Rep tree: PDF version.
Cap tree: PDF version.

9. Collaboratively developing CRESS-GLUE resources.

In this tutorial we have worked toward a particular goal - the creation of a set of bootstrapped phylogenies that can be used to examine the evolutionary history and genome biology of beak and feather disease virus (BFDV).

However, at the same time we created a publicly accessible, sequence-based resource for BFDV - one that can be used repeatedly, for a wide range of comparative studies.

As well as collating the data, we've added some value by cleaning up some of the metadata, and creating an alignment. We've also created some phylogenies that may be of interest to those interested in or impacted by BFDV.

Related Publications