Skip to main content

Lineage prediction from SARS-CoV-2 sequences

Project description

Armadillin

This is an experimental tool under development. The recommended method for calling lineages remains normal Pangolin: https://github.com/cov-lineages/pangolin

A Re-engineered Method Allowing DetermInation of viraL LINeages

Armadillin is an experimental alternative approach to training models on lineages designated by the PANGO team.

Armadillin uses dense neural networks for assignment, which means it doesn't have to assume that positions with an N are the reference sequence. Armadillin is still very fast, in part because it sparsifies the feature input to this neural net during training.

Installation (for inference)

conda create --name armadillin python=3.9
conda activate armadillin
pip3 install armadillin

Usage

You must already have aligned your files to the reference (doing this automatically is on the backlist).

We'll use the COG-UK aligned file for a demo:

wget https://cog-uk.s3.climb.ac.uk/phylogenetics/latest/cog_alignment.fasta.gz
armadillin https://cog-uk.s3.climb.ac.uk/phylogenetics/latest/cog_alignment.fasta.gz

or

armadillin https://cog-uk.s3.climb.ac.uk/phylogenetics/latest/cog_alignment.fasta.gz > output.tsv

Training your own models

Dataset generation

python -m armadillin.training_make_input --designations ~/gisaid/pango-designation-1.2.88/ --gisaid_meta_file ~/gisaid/metadata.tsv --gisaid_mmsa ~/gisaid/msa_2021-10-20.tar.xz --output ~/training_set_nov_02
 python -m armadillin.train --shard_dir /home/theo/training_set_nov_02 --use_wandb --checkpoint_path ~/nov2check1

 python -m armadillin.train --starting_model ~/nov2check1/checkpoint.h5 --use_wandb --checkpoint_path ~/nov2check1_sparse/ --do_pruning --shard_dir /home/theo/training_set_nov_02

 python -m armadillin.training_create_small_model -i /tmp/model_zeros.h5 -d  /home/theo/training_set_nov_02

Related tools

Pangolin is the OG for assigning lineages

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distributions

No source distribution files available for this release.See tutorial on generating distribution archives.

Built Distributions

armadillin-0.44-pp37-pypy37_pp73-win_amd64.whl (81.7 kB view hashes)

Uploaded PyPy Windows x86-64

armadillin-0.44-pp37-pypy37_pp73-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl (189.2 kB view hashes)

Uploaded PyPy manylinux: glibc 2.12+ x86-64 manylinux: glibc 2.5+ x86-64

armadillin-0.44-pp37-pypy37_pp73-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl (189.0 kB view hashes)

Uploaded PyPy manylinux: glibc 2.12+ i686 manylinux: glibc 2.5+ i686

armadillin-0.44-pp37-pypy37_pp73-macosx_10_9_x86_64.whl (174.0 kB view hashes)

Uploaded PyPy macOS 10.9+ x86-64

armadillin-0.44-cp310-cp310-win_amd64.whl (88.4 kB view hashes)

Uploaded CPython 3.10 Windows x86-64

armadillin-0.44-cp310-cp310-win32.whl (76.4 kB view hashes)

Uploaded CPython 3.10 Windows x86

armadillin-0.44-cp310-cp310-musllinux_1_1_x86_64.whl (493.1 kB view hashes)

Uploaded CPython 3.10 musllinux: musl 1.1+ x86-64

armadillin-0.44-cp310-cp310-musllinux_1_1_i686.whl (475.4 kB view hashes)

Uploaded CPython 3.10 musllinux: musl 1.1+ i686

armadillin-0.44-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl (457.6 kB view hashes)

Uploaded CPython 3.10 manylinux: glibc 2.12+ x86-64 manylinux: glibc 2.5+ x86-64

armadillin-0.44-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl (440.6 kB view hashes)

Uploaded CPython 3.10 manylinux: glibc 2.12+ i686 manylinux: glibc 2.5+ i686

armadillin-0.44-cp310-cp310-macosx_10_9_x86_64.whl (188.4 kB view hashes)

Uploaded CPython 3.10 macOS 10.9+ x86-64

armadillin-0.44-cp39-cp39-win_amd64.whl (88.3 kB view hashes)

Uploaded CPython 3.9 Windows x86-64

armadillin-0.44-cp39-cp39-win32.whl (76.4 kB view hashes)

Uploaded CPython 3.9 Windows x86

armadillin-0.44-cp39-cp39-musllinux_1_1_x86_64.whl (491.8 kB view hashes)

Uploaded CPython 3.9 musllinux: musl 1.1+ x86-64

armadillin-0.44-cp39-cp39-musllinux_1_1_i686.whl (473.7 kB view hashes)

Uploaded CPython 3.9 musllinux: musl 1.1+ i686

armadillin-0.44-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl (457.0 kB view hashes)

Uploaded CPython 3.9 manylinux: glibc 2.12+ x86-64 manylinux: glibc 2.5+ x86-64

armadillin-0.44-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl (440.0 kB view hashes)

Uploaded CPython 3.9 manylinux: glibc 2.12+ i686 manylinux: glibc 2.5+ i686

armadillin-0.44-cp39-cp39-macosx_10_9_x86_64.whl (188.5 kB view hashes)

Uploaded CPython 3.9 macOS 10.9+ x86-64

armadillin-0.44-cp38-cp38-win_amd64.whl (88.3 kB view hashes)

Uploaded CPython 3.8 Windows x86-64

armadillin-0.44-cp38-cp38-win32.whl (76.4 kB view hashes)

Uploaded CPython 3.8 Windows x86

armadillin-0.44-cp38-cp38-musllinux_1_1_x86_64.whl (498.7 kB view hashes)

Uploaded CPython 3.8 musllinux: musl 1.1+ x86-64

armadillin-0.44-cp38-cp38-musllinux_1_1_i686.whl (484.0 kB view hashes)

Uploaded CPython 3.8 musllinux: musl 1.1+ i686

armadillin-0.44-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl (466.9 kB view hashes)

Uploaded CPython 3.8 manylinux: glibc 2.12+ x86-64 manylinux: glibc 2.5+ x86-64

armadillin-0.44-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl (451.3 kB view hashes)

Uploaded CPython 3.8 manylinux: glibc 2.12+ i686 manylinux: glibc 2.5+ i686

armadillin-0.44-cp38-cp38-macosx_10_9_x86_64.whl (186.7 kB view hashes)

Uploaded CPython 3.8 macOS 10.9+ x86-64

armadillin-0.44-cp37-cp37m-win_amd64.whl (87.3 kB view hashes)

Uploaded CPython 3.7m Windows x86-64

armadillin-0.44-cp37-cp37m-win32.whl (75.5 kB view hashes)

Uploaded CPython 3.7m Windows x86

armadillin-0.44-cp37-cp37m-musllinux_1_1_x86_64.whl (467.7 kB view hashes)

Uploaded CPython 3.7m musllinux: musl 1.1+ x86-64

armadillin-0.44-cp37-cp37m-musllinux_1_1_i686.whl (454.6 kB view hashes)

Uploaded CPython 3.7m musllinux: musl 1.1+ i686

armadillin-0.44-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl (434.0 kB view hashes)

Uploaded CPython 3.7m manylinux: glibc 2.12+ x86-64 manylinux: glibc 2.5+ x86-64

armadillin-0.44-cp37-cp37m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl (417.6 kB view hashes)

Uploaded CPython 3.7m manylinux: glibc 2.12+ i686 manylinux: glibc 2.5+ i686

armadillin-0.44-cp37-cp37m-macosx_10_9_x86_64.whl (187.5 kB view hashes)

Uploaded CPython 3.7m macOS 10.9+ x86-64

armadillin-0.44-cp36-cp36m-win_amd64.whl (87.3 kB view hashes)

Uploaded CPython 3.6m Windows x86-64

armadillin-0.44-cp36-cp36m-win32.whl (75.5 kB view hashes)

Uploaded CPython 3.6m Windows x86

armadillin-0.44-cp36-cp36m-musllinux_1_1_x86_64.whl (467.4 kB view hashes)

Uploaded CPython 3.6m musllinux: musl 1.1+ x86-64

armadillin-0.44-cp36-cp36m-musllinux_1_1_i686.whl (453.3 kB view hashes)

Uploaded CPython 3.6m musllinux: musl 1.1+ i686

armadillin-0.44-cp36-cp36m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl (433.9 kB view hashes)

Uploaded CPython 3.6m manylinux: glibc 2.12+ x86-64 manylinux: glibc 2.5+ x86-64

armadillin-0.44-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl (417.9 kB view hashes)

Uploaded CPython 3.6m manylinux: glibc 2.12+ i686 manylinux: glibc 2.5+ i686

armadillin-0.44-cp36-cp36m-macosx_10_9_x86_64.whl (187.4 kB view hashes)

Uploaded CPython 3.6m macOS 10.9+ x86-64

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page