get-assemblies 0.4.0

NCBI E-utilities wrapper for assembly downloads

Table of Contents

• Installation

• Dependencies

• Just tell me how to run it

• Overview

• More Examples

• Bugs

• Contributing

• Author Contact

• Acknowledgments

• License

• Copyrights

Installation

get_assemblies is distributed on PyPI as a universal wheel and is available on Linux/macOS and supports Python 3.7+. This software will work on Windows using WSL.

get_assemblies depends on NCBI Entrez Direct which requires Perl. Perl is installed by default on most *nix systems. If edirect is not currently installed, please run get_assemblies --dledirect to install.

$ python3 -m pip install -U get-assemblies

Dependencies

Python modules:

1. python3-wget

2. tqdm

See requirements.txt for more info.

External programs:

1. NCBI Entrez Direct

You can install external programs using the get_assemblies --dledirect command. These will be installed to ${HOME}/edirect unless otherwise specified.

Just tell me how to run it

$ get_assemblies organism 'Pseudomonas fluorescens'

^ This will find all genomes tagged as ‘Pseudomonas fluorescens’ in NCBI’s database. By default, this command will only check to see how many genomes fall into this category.

To download metadata for the genomes:

$ get_assemblies organism 'Pseudomonas fluorescens' --function metadata

Check out the metadata.tab file that is created after running this command. Generally you will want to select a subset from your search. One way to do this is to select the lines that include the genomes of interest, and then saving the assembly accessions to a file. You can either delete the lines that you don’t want, or use grep to pull out the lines that you want to keep. Then you can use cut -f 11 > accs.txt to get the assembly accesions in a file.

$ cat accs.txt | get_assemblies assembly_ids - --function genomes -o fna

This will download the nucleotide fasta files for your genomes of interest.

Overview

This tool was written to make accessing genomic data from NCBI easier. The output files are renamed such that each assembly has a Genus species strain in the filename to make it easy to find the genomes that you’re interested in. You won’t have to spend time renaming the files by hand.

This software is effectively a wrapper for the NCBI edirect tools that makes getting genome files easier. If you are interested in starting a comparative genomics project, this is the tool for you.

The software supports four types of input:

1. organism input, either taxonomy rank names (e.g. Genus species, Family) or taxids

2. assembly ids, either accessions or uids

3. nuccore ids (e.g. individual contig/chromosome names)

4. json input (e.g. the intermediate files - docsums - produced by this script)

Five file type outputs are supported:

1. Nucleotide genome sequence (fna)

2. Nucleotide coding sequence (ffn)

3. Amino acid coding sequence (faa)

4. General feature format (i.e. tab-delimited features) (gff)

5. GenBank format (gbk)

The program will attempt to find a unique prefix per genome assembly. This prefix will be in the resulting filename. A metadata file that contains much of the relevant information per genome will also be included. This file can be included as a supplementary table for a manuscript in a comparative genomics project.

If you need to make phylogenetic trees with these data, check out my other python package, automlsa2.

More Examples

$ get_assemblies organism 'Mycobacterium'
2020-10-15 22:49:53,257 - INFO - Found 7522 genomes to download.
2020-10-15 22:49:53,257 - INFO - Expect 37610MB to 52654MB of data.

$ get_assemblies organism --type ID 167539 --function genomes -o gbk
2020-10-15 23:10:13,822 - INFO - Found 1 genomes to download.
2020-10-15 23:10:13,822 - INFO - Expect 5MB to 7MB of data pending the chosen file types for download.
chunk: 1it [00:01,  1.21s/it]
docsums: 100%|██████████████████████████████| 1/1 [00:00<00:00, 5146.39it/s]
2020-10-15 23:10:16,262 - INFO - Downloading 1 files.
100% [##################################################]           1M / 1M]
2020-10-15 23:10:18,044 - INFO - P_marinus_CCMP1375_SS120.gbk successfully downloaded.
download: 100%|███████████████████████████████| 1/1 [00:01<00:00,  1.78s/it]
$ ls
docsums0.json       metadata.tab
get_assemblies.log  P_marinus_CCMP1375_SS120.gbk

$ echo GCA_000269645.2 | get_assemblies assembly_ids -
2020-10-15 23:18:04,107 - INFO - Found 1 genomes to download.
2020-10-15 23:18:04,107 - INFO - Expect 5MB to 7MB of data pending the chosen file types for download.

Bugs

Viruses are currently not handled well, if at all. Look elsewhere to download those.

Contributing

Feel free to submit bug reports or pull requests so we can improve this software. Undoubtedly there will be some erroneous prefixes generated out there, and I’d like to fix them.

Author Contact

Ed Davis

Acknowledgments

Special thanks for helping me test the software and get the python code packaged:

• Alex Weisberg

• Shawn O’Neil

Also, thanks to these groups for supporting me through my scientific career:

• OSU Chang Lab

• Center for Genome Research and Biocomputing @ OSU

License

get_assemblies is distributed under the terms listed in the LICENSE file. The software is free for non-commercial use.

Copyrights

Copyright (c) 2020 Oregon State University

All Rights Reserved.