`v-build.pl` example usage and command-line options

v-build.pl example usage
v-build.pl command-line options
Building a VADR model library
Advanced tutorial: building an RSV model library
How the VADR 1.0 model library was constructed

`v-build.pl` example usage

v-build.pl creates the VADR model files for a specified reference GenBank (typically RefSeq) sequence necessary for running v-annotate.pl to validate and annotate sequences similar to that reference sequence. It is recommended to run v-build.pl only on sequences of length 25Kb (25,000 nucleotides) or less due to the prohibitively large memory requirements of v-annotate.pl for larger models. To determine the command-line usage of v-build.pl (or any VADR script), use the -h option, like this:

v-build.pl -h

You'll see something like the following output:

# v-build.pl :: build homology model of a single sequence for feature annotation
# VADR 1.6 (Nov 2023)
# - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
# date:    Tue Nov  7 10:07:21 2023
#
Usage: v-build.pl [-options] <accession> <path to output directory to create>

The first few lines are the banner which show the name of the VADR script being run along with the version and release date. This is followed by the time and date the command was executed. The Usage: line details the expected command line arguments. v-build.pl takes as input two command line arguments, the name of the reference accession (<accession>) and the name of the output directory you want it to create (<path to output directory to create>) and populate with output files, including model files for the accession to use with v-annotate.pl.

After that comes a list of all available command-line options. These are explained in more detail below.

Here is an example v-build.pl command using the RefSeq accession NC_039897 a Norovirus GI complete genome sequence, and creating an output directory with the same name as the accession:

v-build.pl NC_039897 NC_039897

If, when you the above command, it fails and you see a (potentially very long) error that starts with Can't locate LWP/Protocol/https.pm in @INC or something similar. You will need to install one or more perl modules before v-build.pl will run as explained more here.

The standard output of v-build.pl that is printed to the screen (which is also output to the .log output file) begins with the banner and date again followed by a list of relevant environment variables, the command line arguments used and any command line options used:

# $VADRBLASTDIR:     /home/nawrocki/vadr-install-dir/ncbi-blast
# $VADREASELDIR:     /home/nawrocki/vadr-install-dir/infernal/binaries
# $VADRINFERNALDIR:  /home/nawrocki/vadr-install-dir/infernal/binaries
# $VADRSCRIPTSDIR:   /home/nawrocki/vadr-install-dir/vadr
#
# accession/model name:  NC_039897
# output directory:      NC_039897

No command line options were used in our example output, but if they were information on them would have appeared after the output directory line.

Next, information is output about each step the script is proceeding through. When each step is completed, the elapsed time for that step is output. The first few steps for v-build.pl are to fetch and parse a FASTA sequence file and feature table file for the input accession NC_039897.

# Fetching FASTA file                                          ... done. [    3.4 seconds]
# Parsing FASTA file                                           ... done. [    0.0 seconds]
# Fetching feature table file                                  ... done. [    3.4 seconds]
# Parsing feature table file                                   ... done. [    0.0 seconds]

Next, any relevant feature tables for proteins referenced in the NC_039897 feature table are fetched and parsed. The fetched data from GenBank is then pruned to remove information that will not be stored in the VADR model files:

# Fetching and parsing protein feature table file(s)           ... done. [   10.3 seconds]
# Pruning data read from GenBank                               ... done. [    0.0 seconds]

The fetched FASTA file is then reformatted to Stockholm for input to cmbuild to create the model CM file, the BLAST nucleotide database is construced, feature information read from the feature tables is finalized internally, and CDS features are translated and used to create the BLAST database with makeblastdb, and the HMMER protein HMM database with hmmbuild.

# Reformatting FASTA file to Stockholm file                    ... done. [    0.0 seconds]
# Building BLAST nucleotide database                           ... done. [    0.7 seconds]
# Finalizing feature information                               ... done. [    0.0 seconds]
# Translating CDS                                              ... done. [    0.2 seconds]
# Building BLAST protein database                              ... done. [    0.3 seconds]
# Building HMMER protein database                              ... done. [    1.1 seconds]
# Checking intron splice sites, if any                         ... done. [    0.0 seconds]

Next, the covariance model is built using cmbuild. This is by far the slowest step of v-build.pl and a rough estimate of how long it will take is output. The longer the sequence being modeled the longer this step will take. When that completes the cm file is pressed with cmpress to prepare it for use with v-annotate.pl, and a BLAST nucleotide database is created from the CM consensus sequence. The final step is to create the model info file:

# Building CM (should take roughly 10-30 minutes)              ... done. [  477.0 seconds]
# Pressing CM file                                             ... done. [    0.4 seconds]
# Creating model info file                                     ... done. [    0.0 seconds]

When all steps are complete, v-build.pl ends by outputting a list of relevant output files with brief descriptions:

# Output printed to screen saved in:                                 NC_039897.vadr.log
# List of executed commands saved in:                                NC_039897.vadr.cmd
# List and description of all output files saved in:                 NC_039897.vadr.filelist
# fasta file for NC_039897 saved in:                                 NC_039897.vadr.fa
# feature table format file for NC_039897 saved in:                  NC_039897.vadr.tbl
# feature table format file for YP_009538340.1 saved in:             NC_039897.vadr.YP_009538340.1.tbl
# feature table format file for YP_009538341.1 saved in:             NC_039897.vadr.YP_009538341.1.tbl
# feature table format file for YP_009538342.1 saved in:             NC_039897.vadr.YP_009538342.1.tbl
# Stockholm alignment file for NC_039897 saved in:                   NC_039897.vadr.stk
# nucleotide blastn db fasta sequence file for NC_039897 saved in:   NC_039897.vadr.fa
# BLAST db .nhr file for NC_039897 saved in:                         NC_039897.vadr.fa.nhr
# BLAST db .nin file for NC_039897 saved in:                         NC_039897.vadr.fa.nin
# BLAST db .nsq file for NC_039897 saved in:                         NC_039897.vadr.fa.nsq
# BLAST db .ndb file for NC_039897 saved in:                         NC_039897.vadr.fa.ndb
# BLAST db .not file for NC_039897 saved in:                         NC_039897.vadr.fa.not
# BLAST db .ntf file for NC_039897 saved in:                         NC_039897.vadr.fa.ntf
# BLAST db .nto file for NC_039897 saved in:                         NC_039897.vadr.fa.nto
# fasta sequence file for CDS from NC_039897 saved in:               NC_039897.vadr.cds.fa
# fasta sequence file for translated CDS from NC_039897 saved in:    NC_039897.vadr.protein.fa
# BLAST db .phr file for NC_039897 saved in:                         NC_039897.vadr.protein.fa.phr
# BLAST db .pin file for NC_039897 saved in:                         NC_039897.vadr.protein.fa.pin
# BLAST db .psq file for NC_039897 saved in:                         NC_039897.vadr.protein.fa.psq
# BLAST db .pdb file for NC_039897 saved in:                         NC_039897.vadr.protein.fa.pdb
# BLAST db .pot file for NC_039897 saved in:                         NC_039897.vadr.protein.fa.pot
# BLAST db .ptf file for NC_039897 saved in:                         NC_039897.vadr.protein.fa.ptf
# BLAST db .pto file for NC_039897 saved in:                         NC_039897.vadr.protein.fa.pto
# HMMER model db file for NC_039897 saved in:                        NC_039897.vadr.protein.hmm
# hmmbuild build output (concatenated) saved in:                     NC_039897.vadr.protein.hmmbuild
# binary HMM and p7 HMM filter file saved in:                        NC_039897.vadr.protein.hmm.h3m
# SSI index for binary HMM file saved in:                            NC_039897.vadr.protein.hmm.h3i
# optimized p7 HMM filters (MSV part) saved in:                      NC_039897.vadr.protein.hmm.h3f
# optimized p7 HMM filters (remainder) saved in:                     NC_039897.vadr.protein.hmm.h3p
# hmmpress output file saved in:                                     NC_039897.vadr.hmmpress
# CM file saved in:                                                  NC_039897.vadr.cm
# cmbuild output file saved in:                                      NC_039897.vadr.cmbuild
# binary CM and p7 HMM filter file saved in:                         NC_039897.vadr.cm.i1m
# SSI index for binary CM file saved in:                             NC_039897.vadr.cm.i1i
# optimized p7 HMM filters (MSV part) saved in:                      NC_039897.vadr.cm.i1f
# optimized p7 HMM filters (remainder) saved in:                     NC_039897.vadr.cm.i1p
# cmpress output file saved in:                                      NC_039897.vadr.cmpress
# VADR 'model info' format file for NC_039897 saved in:              NC_039897.vadr.minfo
#
# All output files created in directory ./NC_039897/

These files include the FASTA and and their formats are described more here.

Only some of these files will be used by v-annotate.pl. These are the files with the following suffixes:

file suffix	description	reference
`.protein.fa.p{hr,in,sq,db,ot,tf,to}`	BLAST protein database index files, created by `makeblastdb`	binary files, not meant to be human-readable
`.fa.n{hr,in,sq,db,ot,tf,to}`	BLAST nucleotide database index files, created by `makeblastdb`	binary files, not meant to be human-readable
`.cm`	Infernal 1.1x covariance model file	http://eddylab.org/infernal/Userguide.pdf (section 9: "File and output formats")
`.cm.i1{m,i,f,p}`	Infernal 1.1x covariance model index files, created by `cmpress`	binary files, not meant to be human-readable
`.hmm`	HMMER 3.x HMM file	http://eddylab.org/software/hmmer/Userguide.pdf ("HMMER profile HMM files" section)
`.hmm.h3{m,i,f,p}`	HMMER 3.x covariance model index files, created by `hmmpress`	binary files, not meant to be human-readable
`.minfo`	VADR model info file	description of format

You can use only this model to annotate sequences with v-annotate.pl that are similar to NC_039897 using the --mdir and --mkey options with a command like v-annotate.pl --mdir NC_039897 --mkey NC_039897.vadr <fasta-seq-file> <output-directory>, explained more here.

Alternatively, you can combine these files together with analogous files from additional v-build.pl runs for other accessions to create a VADR model library and then use v-annotate.pl --mdir and --mkey options to specify that the library be used. This is explained in more detail below. The VADR 1.0 library was created in this manner, as explained in another section.

`v-build.pl` command-line options

To get a list of command-line options, execute:

v-build.pl -h

This will output the usage and available command-line options. Each option has a short description, but additional information on some of these options can be found below. For v-build.pl the available options are split into eight different categories, each explained in their own subsection below.

In the tables describing options below, <s> represents a string and <n> represents an integer.

`v-build.pl` basic options

The first category of options are the basic options:

........option........	explanation
`-f`	if `<output directory>` already exists, then using this option will cause it to be overwritten, otherwise the progam exits in error
`-v`	verbose mode: all commands will be output to standard output as they are run
`--stk <s>`	single sequence stockholm alignment will be read from `<s>`, possibly with secondary structure annotation
`--infa <s>`	instead of fetching the FASTA for this sequence from GenBank, use the sequence in file
`--inft <s>`	instead of fetching the feature table for this sequence from GenBank, use the feature table in file
`--ftfetch1`	use `efetch` program (must be in your `PATH`) to fetch feature table with `efetch -format ft` instead of default method of fetching from an `eutils` URL
`--ftfetch2`	use `efetch` program (must be in your `PATH`) to fetch feature table with `efetch -format gbc
`--gb`	fetch and parse a GenBank-format file from GenBank instead of a feature table
`--ingb <s>`	read the GenBank-format file in `<s>` instead of a feature table file (requires `--gb`)
`--addminfo <s>`	add arbitrary feature info in file `<s>` to output `.minfo` file, see an example here
`--forcelong`	use at your own risk; allow long models > 25Kb in length; by default `v-build.pl` will fail for any model more than 25Kb (25,000 nucleotides) because `v-build.pl` will be very slow and the memory requirements of `v-annotate.pl` will be prohibitively large
`--keep`	keep additional output files that are normally removed

`v-build.pl` options for controlling what feature types are stored in the output model info file

By default, only CDS, gene and mat_peptide feature types read from the GenBank feature table file will be stored in the output .minfo file. This default set can be changed using the following three command line options. For an example of using the --fadd option, see the construction of the dengue virus RefSeq models for the VADR 1.0 model library here.

......option......	explanation
`--fall`	specify that all feature types (except those in `<s>` from `--fskip <s>`) be added to the `.minfo` output file
`--fadd <s>`	add feature types listed in `<s>` to the default set, where `<s>` is a comma-separated string with each feature type separated by a comma with no whitespace
`--fskip <s>`	do not store information for feature types listed in `<s>`, where `<s>` is a comma-separated string with each feature type separated by a comma with no whitespace; `<s>` may contain feature types from the default set, or from other features (if `--fall` also used)

`v-build.pl` options for controlling what qualifiers are stored in the output model info file

By default, only product, gene and exception qualifiers read from the GenBank feature table file will be stored in the output .minfo file, and then only for the feature types that will be stored. This default set can be changed using the following five command line options. For an example of using the --qadd and --qftradd options, see the construction of the dengue virus RefSeq models for the VADR 1.0 model library here.

.......option.......	explanation
`--qall`	specify that all qualifiers (except those in `<s>` from `--qskip <s>`) be added to the `.minfo` output file
`--qadd <s>`	add qualifiers listed in `<s>` to the default set, where `<s>` is a comma-separated string with each qualifier separated by a comma with no whitespace
`--qftradd <s>`	specify that the qualifiers listed in `<s2>` from `qadd <s2>` only apply for feature types in the string `<s>`, where `<s>` is a comma-separated string with each qualifier separated by a comma with no whitespace
`--qskip <s>`	do not store information for qualifiers listed in `<s>`, where `<s>` is a comma-separated string with each qualifier separated by a comma with no whitespace; `<s>` may contain qualifiers from the default set, or from other qualifiers (if `--qall` also used)
`--noaddgene`	do not automatically add `gene` qualifiers from `gene` features to any overlapping non-gene features
`--nosplice`	do not automatically check for GT/AG intron splice sites and add `canon_splice_sites` qualifiers for introns with valid splice sites, an intron is defined as a gap between CDS segments >= `<n>` nucleotides from `--intlen` option, by default `<n>` is `40`
`--ssplice`	exit if any introns in CDS are found that do not have valid GT/AG splice sites, introns defined as explained in `--nosplice` description above

`v-build.pl` options for including additional model attributes

Besides qualifiers read from GenBank and information included in the input .minfo file with the --addminfo option, two additional attributes can be added using the command line options --group and --subgroup. For an example of using these options see construction of the Norovirus VADR 1.0 library model files here.

.....option.....	explanation
`--group <s>`	specify that the model `group` attribute is `<s>`, e.g. `Norovirus`
`--subgroup <s>`	specify that the model `subgroup` attribute is `<s>`, e.g. `GI`, requires `--group`

`v-build.pl` options for controlling the CDS translation step

By default, the NCBI translation table 1 is used to translate CDS sequences into proteins. This can be changed to use translation table <s> with the --ttbl option. A non-1 translation table will be stored in the output .minfo file so that v-annotate.pl is aware of the translation table to use when analyzing CDS predictions.

......option......	explanation
`--ttbl <n>`	specify that NCBI translation table `<n>` be used instead of `1`

Reference on NCBI translation tables: https://www.ncbi.nlm.nih.gov/Taxonomy/taxonomyhome.html/index.cgi?chapter=cgencodes

`v-build.pl` options for controlling the `cmbuild` step

Several options exist for controlling the command-line options that will be passed to Infernal's cmbuild program. For more information on these options and how they control cmbuild, see the Infernal User's Guide manual page for cmbuild (section 8 of http://eddylab.org/infernal/Userguide.pdf) .

......option......	explanation
`--cmn <n>`	set the number of seqs for glocal forward profile HMM filter calibration to `<n>` (sets the `cmbuild --EgfN` option), default is to use default `cmbuild` value
`--cmp7ml`	set CM's filter profile HMM as the maximum likelihood profile HMM (sets the `cmbuild --p7ml` option)
`--cmere`	set CM relative entropy target bits to position to `<x>` (sets the `cmbuild --ere <x>` option), default is to use default `cmbuild` value
`--cmeset`	set CM effective sequence number to '`(sets the`cmbuild --eset `option), default is to use default`cmbuild` value
`--cmemaxseq`	set CM maximum allowed effective sequence # for CM to `<x>` (sets the `cmbuild --emaxseq <x>` option)
`--cmnoh3pri`	do not use `--noh3pri` option with `cmbuild`, to allow CM to use prior from HMMER3 for zero basepair models
`--cminfile`	read `cmbuild` options from an input file `<s>`, the contents of the file (after removing newlines) will be supplied directly to `cmbuild` as an options string (possibly with more than one option separated by whitespace)

`v-build.pl` options for skipping stages

......option......	explanation
`--skipbuild`	skip the `cmbuild` step; this is mostly useful for debugging purposes, but also possibly for creating a different `.minfo` file using options like `--fadd` and/or `--qadd` for an already created model without the need to wait for the slow `cmbuild` step
`--onlyurl`	output the url for the GenBank feature table file (or GenBank-format file if `--gb` also used, and exit; possibly helpful if `v-build.pl` is having trouble fetching from GenBank, you can manually download the feature table file from a browswer with the output url and use the downloaded file with `--inft`

`v-build.pl` options for optional output files

.......option.......	explanation
`--ftrinfo <s>`	output information on the internal data structure used to keep track of features to `<s>`, mainly useful for debugging
`--sgminfo <s>`	output information on the internal data structure used to keep track of segments of features to `<s>`, mainly useful for debugging

Other `v-build.pl` expert options

.......option.......	explanation
`--execname <s>`	in banner and usage output, replace `v-annotate.pl` with `<s>`
`--nosig2mat`	do not treat `sig_peptide` features as `mat_peptide` in model info file
`--intlen <n>`	define intron as any gap >= `<n>` nucleotides between segments in a CDS, only relevant for checking for canonical splice sites, the default value for `<n>` is `40`

Building a VADR model library

Follow these steps to build a VADR model library:

Run v-build.pl multiple (N>1) times for different accessions. If your model is larger than 30Kb, or you know that you will use the -s and --glsearch options with v-annotate.pl as is recommended for coronavirus annotation (https://github.com/ncbi/vadr/wiki/Coronavirus-annotation), you should also use the --skipbuild option with v-build.pl. If you do, no CM files will be created so you can skip steps 4 and 8 below and ignore any other steps involving a .cm file.
Create a new directory call it my-vadr-model-dir, for example.
Concatenate all resulting N .vadr.minfo files into a single file, call it my.vadr.minfo, and move it to the my-vadr-model-dir directory.
Concatenate all resulting N .vadr.cm files into a single file, call it my.vadr.cm, for example, and move it to the my-vadr-model-dir directory.
Concatenate all resulting N .vadr.hmm files into a single file, call it my.vadr.hmm, for example, and move it to the my-vadr-model-dir directory.
Concatenate all resulting N .vadr.fa files into a single file, call it my.vadr.fa, for example, and move it to the my-vadr-model-dir directory, and then create a BLAST nucleotide database from it with the command:

$VADRBLASTDIR/makeblastdb -dbtype nucl -in my-vadr-model-dir/my.vadr.fa

Create an Easel index .ssi file for the .vadr.fa file you created and moved in step 6 with the command:

$VADREASELDIR/esl-sfetch --index my-vadr-model-dir/my.vadr.fa

Move all resulting BLAST protein DB files (.vadr.protein.fa, .vadr.protein.fa.p{hr,in,sq,db,ot,tf,to}) from all N runs into the my-vadr-model-dir directory.
Run cmpress on the my.vadr.cm file created in step 4 like this:

$VADRINFERNALDIR/cmpress my-vadr-model-dir/my.vadr.cm

Run hmmpress on the my.vadr.hmm file created in step 5 like this:

$VADRHMMERDIR/hmmpress my-vadr-model-dir/my.vadr.hmm

You can then use this library with v-annotate.pl as follows:

v-annotate.pl --mdir my-vadr-model-dir --mkey my.vadr <fasta-file>
<output directory>

Or substitute the full paths to my-vadr-model-dir if it is not a subdirectory of the current directory. Optionally, you can supply the paths to each of the relevant files or directories, possibly if they are in different directories or are not consistently named, using the -m, -a, -i, -n, and -x options as listed here, with a command like:

v-annotate.pl -m my-vadr-model-dir/my.vadr.cm -a my-vadr-model-dir/my.vadr.hmm -i my-vadr-model-dir/my.vadr.minfo -n
my-vadr-model-dir/my.vadr.fa -x my-vadr-model-dir <fasta-file> <output directory>

If you used --skipbuild with v-build.pl, you will also have to use the -s and --glsearch options with v-annotate.pl.

If you ever move .cm, .hmm, or BLAST .fa files into new directories, make sure you also move the corresponding index files (.cm.*, .hmm.*, .fa.*) along with them.

Advanced tutorial: building an RSV model library (link to different page)

How the VADR 1.0 model library was constructed

The VADR 1.0 library was built with version 1.0 of VADR. It has not changed since version 1.0 and is still the default model library used in this version. It is included here as an example of how to build a VADR library, but also so it can be reproduced, because it is the model library used in the paper on VADR 1.0 (https://doi.org/10.1186/s12859-020-3537-3). If you want to reproduce it exactly, you'll want to install version 1.0. The install script for v1.0 is here:

https://raw.githubusercontent.com/ncbi/vadr/vadr-1.0/vadr-install.sh

Additionally, if the RefSeq annotation for any of these 197 VADR models has changed since October 2019, then it may not be able to identically reproduce the VADR 1.0 model library using the steps outlined above. This is because v-build.pl fetches the current RefSeq annotation data from GenBank when it is run. If necessary, email eric.nawrocki@nih.gov for additional files needed to reproduce the library exactly.

Also note that the library has changed since version 1.0. For example, the default set of models included with version 1.1 has 205 total models, not 197. To see a list of changes, see the RELEASE-NOTES.txt file in the directory pointed to by your $VADRMODELDIR environment variable after installing VADR. To reproduce the construction of the 1.1 library, you would run similar steps to those below but also adding the additional models listed in the RELEASE-NOTES.txt file.

Additionally, as of version 1.2, the Caliciviridae models and Flaviviridae models have been split up into two different model sets, but both are installed by the VADR install script vadr-install.sh. The Caliciviridae models are used by default with v-annotate.pl. To use the Flaviviridae models, use the options --mkey flavi --mdir $VADRMODELDIR/vadr-models-flavi.

Models in the VADR 1.0 library

The VADR 1.0 model library consists of 197 VADR models. Nine of these are Norovirus RefSeq models, listed in vadr/documentation/build-files/1p0-models/norovirus.9.list. Four of these are Dengue virus RefSeq models, listed in vadr/documentation/build-files/1p0-models/dengue.4.list. Eight of these are Hepatitis C virus Refseq models, listed in vadr/documentation/build-files/1p0-models/hcv.8.list. The remaining 173 are additional Caliciviridae and Flaviviridae RefSeq models, listed in vadr/documentation/build-files/1p0-models/non-noro-dengue-hcv.173.list.

Building the VADR 1.0 library Norovirus models

To build models for each of the nine norovirus RefSeqs listed in vadr/documentation/build-files/1p0-models/norovirus.9.list, run v-build.pl nine separate times as follows:

v-build.pl --group Norovirus --subgroup GI NC_001959 NC_001959
v-build.pl --group Norovirus --subgroup GV NC_008311 NC_008311
v-build.pl --group Norovirus --subgroup GIII NC_029645 NC_029645
v-build.pl --group Norovirus --subgroup GII --addminfo NC_029646.addminfo NC_029646 NC_029646
v-build.pl --group Norovirus --subgroup GIV NC_029647 NC_029647
v-build.pl --group Norovirus --subgroup GI NC_031324 NC_031324
v-build.pl --group Norovirus --subgroup GII --addminfo NC_039475.addminfo NC_039475 NC_039475
v-build.pl --group Norovirus --subgroup GII NC_039476 NC_039476
v-build.pl --group Norovirus --subgroup GII NC_039477 NC_039477

(The shell script vadr/documentation/build-files/1p0-models/norovirus.9.build.sh contains these commands.)

The --group and --subgroup options specify the group and subgroup values that will be added to the output .minfo file. These will enable v-annotate.pl to label sequences classified to these models with the corresponding group and subgroup, as well as to fail sequences users expect are norovirus sequences that actually are classified best to a non-norovirus model.

The --addminfo options specify an input file that contains information on additional feature attributes not from GenBank that are desired in the output .minfo file. Those input files are located here: vadr/documentation/build-files/1p0-models/NC_029646.addminfo and vadr/documentation/build-files/1p0-models/NC_039475.addminfo. The NC_029646.addminfo file is in the .minfo format and looks like this:

MODEL NC_029646 
FEATURE NC_029646 type:"CDS" coords:"5085..6692:+" xmaxins_exc:"297:36"

This file specifies that the additional <key>:<value> pair of xmaxins_exc:"297:36" be added to the CDS feature with coordinates 5085..6692:+. (The VADR coordinate string format is described here). The NC_039475.addminfo file is similar except with the value "295:36". These two additions allow the corresponding CDS features to have an exception to the default maximum allowed insert length by v-annotate.pl without causing an insertnp alert, setting it as 36 after position 297 in NC_029646 and after position 295 in NC_039475. This change to the default was allowed after GenBank indexers observed a common biologically valid insertion at these positions of length 36 nucleotides (nt), which exceeds the default maximum of 27 nt.

Building the VADR 1.0 library Dengue virus models

The four Dengue RefSeq models are built using the --stk option to specify the secondary structure of structured regions of the genome at the 5' and 3' ends, as well as some additional options. To build models for each of these RefSeqs, run run v-build.pl from VADR v1.0 four separate times as follows:

v-build.pl --stk NC_001477.v1.stk --qftradd stem_loop,ncRNA --qadd note,ncRNA_class --fadd stem_loop,ncRNA --group Dengue --subgroup 1 NC_001477 NC_001477
v-build.pl --stk NC_001474.v1.stk --qftradd stem_loop,ncRNA --qadd note,ncRNA_class --fadd stem_loop,ncRNA --group Dengue --subgroup 2 NC_001474 NC_001474
v-build.pl --stk NC_001475.v1.stk --qftradd stem_loop,ncRNA --qadd note,ncRNA_class --fadd stem_loop,ncRNA --group Dengue --subgroup 3 NC_001475 NC_001475
v-build.pl --stk NC_002640.v1.stk --qftradd stem_loop,ncRNA --qadd note,ncRNA_class --fadd stem_loop,ncRNA --group Dengue --subgroup 4 NC_002640 NC_002640

(The shell script vadr/documentation/build-files/1p0-models/dengue.4.build.sh contains these commands.)

The --qftradd, --qadd, and --fadd options all take comma-separated strings as arguments. For example, --qftradd takes the argument stem_loop,ncRNA. These options specify that v-build.pl should include GenBank feature information for stem_loop and ncRNA features in addition to its default set of CDS, gene, and mat_peptidefeatures. Further, thenoteandncRNA_classGenBank qualifiers should be included in addition to the default set ofproduct, geneandexceptionqualifiers for thesestem_loopandncRNA` features.

The --group and --subgroup options are used in a similar way to how they were used to build the norovirus models.

Manually edit the `.minfo` files for the Dengue models

Not all the annotation in the Dengue RefSeqs as of October 2019 is in the VADR 1.0 library. To make your .minfo files match those in the library exactly do the following:

Remove the ncRNA features with product names sfRNA2, sfRNA3, and sfRNA4 from the .minfo files of NC_001477, NC_001474, and NC_001475.
Remove the ncRNA features with product names sfRNA3 and sfRNA4 from the .minfo file of NC_002640.
Remove all occurrences of the substring gene:"-" (only) from FEATURE lines in the .minfo files of NC_001477, NC_001474, NC_001475 and NC_002640.

Building the VADR 1.0 library Hepatitis C virus models

To build models for each of the eight Hepatitis C RefSeqs listed in vadr/documentation/build-files/1p0-models/hcv.8.list, run v-build.pl eight separate times as follows:

v-build.pl --group HCV --subgroup 1 NC_004102 NC_004102
v-build.pl --group HCV --subgroup 1 NC_038882 NC_038882
v-build.pl --group HCV --subgroup 2 NC_009823 NC_009823
v-build.pl --group HCV --subgroup 3 NC_009824 NC_009824
v-build.pl --group HCV --subgroup 4 NC_009825 NC_009825
v-build.pl --group HCV --subgroup 5 NC_009826 NC_009826
v-build.pl --group HCV --subgroup 6 NC_009827 NC_009827
v-build.pl --group HCV --subgroup 7 NC_030791 NC_030791

(The shell script vadr/documentation/build-files/1p0-models/hcv.8.build.sh contains these commands.)

The --group and --subgroup options are used in a similar way to how they were used to build the norovirus models.

Building the VADR 1.0 library models for the other 173 Caliciviridae and Flaviviridae viral species

To build models for the other 173 Caliciviridae and Flaviviridae models listed in vadr/documentation/build-files/1p0-models/non-noro-dengue-hcv.173.list

Simply run v-build.pl from VADR v1.0 using default parameters for each accession. For example:

  v-build.pl NC_034444 NC_034444

(The shell script vadr/documentation/build-files/1p0-models/non-noro-dengue.hcv.173.build.sh will execute these 173 commands.)

Each of these commands takes roughtly between 10 minutes and an hour.

Concatenating files for the 197 models to create the VADR library

After completing the steps above to make the 197 models, you can make the VADR 1.0 library by following the instructions for creating a VADR library starting at step 2 here.

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

build.md

build.md

`v-build.pl` example usage and command-line options