geoseeq 0.7.13

GeoSeeq command line tools and python API

Geoseeq API Client

This package is a python library to interact with a Geoseeq server. It includes a command line interface that may be used to perform common tasks GeoSeeq tasks from the terminal.

GeoSeeq is a platform for sharing biological, climatological, and public health datasets. Learn more here.

This API client is a work in progress and we welcome suggestions, feedback, comments, and criticisms.

---

Installation

Install from PyPi

pip install geoseeq

Install from source

Download this directory and run python setup.py install

---

Using the Command Line

Run the command line by typing geoseeq into a terminal prompt. See available options by adding --help

$ geoseeq --help

Configuration and Using an API token

For many tasks you will need an API token to interact with GeoSeeq. You can get this token by logging into the GeoSeeq Portal going to your user profile and clicking the "Tokens" tab.

Once you have a token you will need to configure GeoSeeq to use it. Run geoseeq config and leave the profile name and url blank. You will be prompted to enter your API token.

$ geoseeq config
Set custom profile name? (Leave blank for default) []:
Enter the URL to use for GeoSeeq (Most users can use the default) [https://backend.geoseeq.com]:
Enter your GeoSeeq API token:
Profile configured.

This command will store your token in a file called ~/.config/geoseeq/profiles.json and will be used by all future commands.

Example Commands

You can find more command line examples in docs/

Download Short Read Sequencing data from one sample in a project as a set of FASTQ files

This command will download data from this project.

$ geoseeq download files --extension fastq.gz "GeoSeeq/Example CLI Project"

Uploading sequencing data

GeoSeeq can automatically group fastq files into samples according to their sample name, read number, and lane number. It supports paired end, single end, nanopore, and pacbio reads.

Assume you have data from a single ended sequencing run stored as fastq files:

• Sample1_L1_R1.fastq.gz
• Sample1_L1_R2.fastq.gz
• Sample1_L2_R1.fastq.gz
• Sample1_L2_R2.fastq.gz

You can upload these files to GeoSeeq using the command line. This example will upload 32 files in parallel:

# navigate to the directory where the fastq files are stored
$ ls -1 *.fastq.gz > fastq_files.txt  # check that files are present

$ geoseeq upload reads --cores 32 "GeoSeeq/Example CLI Project" fastq_files.txt
Using regex: "(?P<sample_name>[^_]*)_L(?P<lane_num>[0-9]*)_R(?P<pair_num>1|2)\.fastq\.gz"
All files successfully grouped.
sample_name: Sample1
  module_name: short_read::paired_end
    short_read::paired_end::read_1::lane_1: Sample1_L1_R1.fastq.gz
    short_read::paired_end::read_2::lane_1: Sample1_L1_R2.fastq.gz
    short_read::paired_end::read_1::lane_2: Sample1_L2_R1.fastq.gz
    short_read::paired_end::read_2::lane_2: Sample1_L2_R2.fastq.gz
Do you want to upload these files? [y/N]: y
Uploading Sample: Sample1

GeoSeeq will automatically create a new sample named Sample1 if it does not already exist.

This command would upload data to this project.. Since only organization members can upload data, you will need to replace GeoSeeq with your organization name.

To rename samples on the fly, provide a CSV file with current and new names using the --name-map option:

$ geoseeq upload reads --name-map sample_map.csv current_name new_name "GeoSeeq/Example CLI Project" fastq_files.txt

Note: You will need to have an API token set to use this command (see above)

If your reads already live in an S3-compatible bucket (AWS S3, Backblaze B2, Wasabi, MinIO, ...) and you want GeoSeeq to point at them in place rather than upload the bytes, see Linking remote files below.

Linking remote files

If your fastq files already exist in an S3-compatible bucket — AWS S3, Backblaze B2, Wasabi, MinIO, etc. — you can register them with GeoSeeq as links instead of uploading the bytes. Pick the right command for the shape of the work:

Command	Use when
geoseeq link reads	You have a whole S3 prefix of fastq files to register in bulk. GeoSeeq lists the prefix, groups files into samples server-side, and links each file in place.
geoseeq upload reads	You want GeoSeeq to manage the bytes — local files copied into managed storage.
geoseeq s3 register	You have a single file already staged in GeoSeeq's per-project staging credentials flow.

link reads requires the optional [s3] extras (boto3):

pip install 'geoseeq[s3]'

Example — register a Backblaze B2 prefix of paired-end reads:

$ geoseeq link reads "MyOrg/MyProject" \
      s3://my-bucket/path/to/reads/ \
      --endpoint-url https://s3.us-east-005.backblazeb2.com \
      --commit

Without --commit the command runs in dry-run mode and prints the proposed (sample, field, s3-uri) tuples without writing. AWS credentials follow the standard boto3 chain (env vars, shared config, instance profile) — there are no new secrets for the CLI to manage.

Note: geoseeq upload reads --link-type s3 <local-file-list> is deprecated as of this release; prefer geoseeq link reads (bulk) or geoseeq s3 register (single staged file).

Using the Python API in a program

Please see geoseeq_api/cli/download.py for examples of how to download data using the Python API directly.

Development in GitHub Codespaces

This repository includes a .devcontainer configuration so it can be opened directly in GitHub Codespaces. When the codespace is created the development dependencies are installed and pre-commit hooks are set up automatically.

---

Notes

Terminology

Some terms have changed in GeoSeeq since this package was written. The command line tool and code may contain references to old names.

Old Name	New Name
Sample Group	Project
Library	defunct
Analysis Result	ResultFolder
Analysis Result Field	ResultFile

---

License and Credits

GeoSeeq is built and maintained by Biotia

The GeoSeeq API client is licensed under the MIT license.