SARS-Cov-2 Recombinant Finder for fasta sequences

Overview

Sc2rf - SARS-Cov-2 Recombinant Finder

Pronounced: Scarf

What's this?

Sc2rf can search genome sequences of SARS-CoV-2 for potential recombinants - new virus lineages that have (partial) genes from more than one parent lineage.

Is it already usable?

This is a very young project, started on March 5th, 2022. As such, proceed with care. Results may be wrong or misleading, and with every update, anything can still change a lot.

Anyway, I'm happy that scientists are already seeing benefits from Sc2rf and using it to prepare lineage proposals for cov-lineages/pango-designation.

Though I already have a lot of ideas and plans for Sc2rf (see at the bottom of this document), I'm very open for suggestions and feature requests. Please write an issue, start a discussion or get in touch via mail or twitter!

Example output

Screenshot of the terminal output of Sc2rf

Requirements and Installation

You need at least Python 3.6 and you need to install the requirements first. You might use something like python3 -m pip install -r requirements.txt to do that. There's a setup.py which you should probably ignore, since it's work in progress and does not work as intented yet.

Also, you need a terminal which supports ANSI control sequences to display colored text. On Linux, MacOS, etc. it should probably work.

On Windows, color support is tricky. On a recent version of Windows 10, it should work, but if it doesn't, install Windows Terminal from GitHub or Microsoft Store and run it from there.

Basic Usage

Start with a .fasta file with one or more sequences which might contain recombinants. Your sequences have to be aligned to the reference.fasta. If they are not, you will get an error message like:

Sequence hCoV-19/Phantasialand/EFWEFWD not properly aligned, length is 29718 instead of 29903.

(For historical reasons, I always used Nextclade to get aligned sequences, but you might also use Nextalign or any other tool. Installing them is easy on Linux or MacOS, but not on Windows. You can also use a web-based tool like MAFFT.)

Then call:

sc2rf.py <your_filename.fasta>

If you just need some fasta files for testing, you can search the pango-lineage proposals for recombinant issues with fasta-files, or take some files from my shared-sequences repository, which might not contain any actual recombinants, but hundreds of sequences that look like they were!

No output / some sequences not shown

By default, a lot filters are active to show only the likely recombinants, so that you can input 10000s of sequences and just get output for the interesting ones. If you want, you can disable all filters like that, which is only recommended for small input files with less than 100 sequences:

sc2rf.py --parents 1-35 --breakpoints 0-100 \
--unique 1 --max-ambiguous 10000 <your_filename.fasta>

or even

sc2rf.py --parents 1-35 --breakpoints 0-100 \
--unique 1 --max-ambiguous 10000 --force-all-parents \
--clades all <your_filename.fasta>

The meaning of these parameters is described below.

Advanced Usage

You can execute sc2rf.py -h to get excactly this help message:

usage: sc2rf.py [-h] [--primers [PRIMER ...]]
                [--primer-intervals [INTERVAL ...]]
                [--parents INTERVAL] [--breakpoints INTERVAL]
                [--clades [CLADES ...]] [--unique NUM]
                [--max-intermission-length NUM]
                [--max-intermission-count NUM]
                [--max-name-length NUM] [--max-ambiguous NUM]
                [--force-all-parents]
                [--select-sequences INTERVAL]
                [--enable-deletions] [--show-private-mutations]
                [--rebuild-examples] [--mutation-threshold NUM]
                [--add-spaces [NUM]] [--sort-by-id [NUM]]
                [--verbose] [--ansi] [--hide-progress]
                [--csvfile CSVFILE]
                [input ...]

Analyse SARS-CoV-2 sequences for potential, unknown recombinant
variants.

positional arguments:
  input                 input sequence(s) to test, as aligned
                        .fasta file(s) (default: None)

optional arguments:
  -h, --help            show this help message and exit

  --primers [PRIMER ...]
                        Filenames of primer set(s) to visualize.
                        The .bed formats for ARTIC and EasySeq
                        are recognized and supported. (default:
                        None)

  --primer-intervals [INTERVAL ...]
                        Coordinate intervals in which to
                        visualize primers. (default: None)

  --parents INTERVAL, -p INTERVAL
                        Allowed number of potential parents of a
                        recombinant. (default: 2-4)

  --breakpoints INTERVAL, -b INTERVAL
                        Allowed number of breakpoints in a
                        recombinant. (default: 1-4)

  --clades [CLADES ...], -c [CLADES ...]
                        List of variants which are considered as
                        potential parents. Use Nextstrain clades
                        (like "21B"), or Pango Lineages (like
                        "B.1.617.1") or both. Also accepts "all".
                        (default: ['20I', '20H', '20J', '21I',
                        '21J', 'BA.1', 'BA.2', 'BA.3'])

  --unique NUM, -u NUM  Minimum of substitutions in a sample
                        which are unique to a potential parent
                        clade, so that the clade will be
                        considered. (default: 2)

  --max-intermission-length NUM, -l NUM
                        The maximum length of an intermission in
                        consecutive substitutions. Intermissions
                        are stretches to be ignored when counting
                        breakpoints. (default: 2)

  --max-intermission-count NUM, -i NUM
                        The maximum number of intermissions which
                        will be ignored. Surplus intermissions
                        count towards the number of breakpoints.
                        (default: 8)

  --max-name-length NUM, -n NUM
                        Only show up to NUM characters of sample
                        names. (default: 30)

  --max-ambiguous NUM, -a NUM
                        Maximum number of ambiguous nucs in a
                        sample before it gets ignored. (default:
                        50)

  --force-all-parents, -f
                        Force to consider all clades as potential
                        parents for all sequences. Only useful
                        for debugging.

  --select-sequences INTERVAL, -s INTERVAL
                        Use only a specific range of input
                        sequences. DOES NOT YET WORK WITH
                        MULTIPLE INPUT FILES. (default: 0-999999)

  --enable-deletions, -d
                        Include deletions in lineage comparision.

  --show-private-mutations
                        Display mutations which are not in any of
                        the potential parental clades.

  --rebuild-examples, -r
                        Rebuild the mutations in examples by
                        querying cov-spectrum.org.

  --mutation-threshold NUM, -t NUM
                        Consider mutations with a prevalence of
                        at least NUM as mandatory for a clade
                        (range 0.05 - 1.0, default: 0.75).

  --add-spaces [NUM]    Add spaces between every N colums, which
                        makes it easier to keep your eye at a
                        fixed place. (default without flag: 0,
                        default with flag: 5)

  --sort-by-id [NUM]    Sort the input sequences by the ID. If
                        you provide NUM, only the first NUM
                        characters are considered. Useful if this
                        correlates with meaning full meta
                        information, e.g. the sequencing lab.
                        (default without flag: 0, default with
                        flag: 999)

  --verbose, -v         Print some more information, mostly
                        useful for debugging.

  --ansi                Use only ASCII characters to be
                        compatible with ansilove.

  --hide-progress       Don't show progress bars during long
                        task.

  --csvfile CSVFILE     Path to write results in CSV format.
                        (default: None)

An Interval can be a single number ("3"), a closed interval
("2-5" ) or an open one ("4-" or "-7"). The limits are inclusive.
Only positive numbers are supported.

Interpreting the output

To be written...

There already is a short Twitter thread which explains the basics.

Source material attribution

  • virus_properties.json contains data from LAPIS / cov-spectrum which uses data from NCBI GenBank, prepared and hosted by Nextstrain, see blog post.
  • reference.fasta is taken from Nextstrain's nextclade_data, see NCBI for attribution.
  • mapping.csv is a modified version of the table on the covariants homepage by Nextstrain.
  • Example output / screenshot based on Sequences published by the German Robert-Koch-Institut.
  • Primers:
    • ARTIC primers CC-BY-4.0 by the ARTICnetwork project
    • EasySeq primers by Coolen, J. P., Wolters, F., Tostmann, A., van Groningen, L. F., Bleeker-Rovers, C. P., Tan, E. C., ... & Melchers, W. J. Removed until I understand the format if the .bed file. There will be an issue soon.
    • midnight primers CC-BY-4.0 by Silander, Olin K, Massey University

The initial version of this program was written in cooperation with @flauschzelle.

TODO / IDEAS / PLANS

  • Move these TODOs into actual issues
  • add disclaimer and link to pango-designation
  • provide a sample file (maybe both .fasta and .csv, as long as the csv step is still needed)
  • accept aligned fasta
    • as input file
    • as piped stream
  • If we still accept csv/ssv input, autodetect the delimiter either by file name or by analysing the first line
  • find a way to handle already designated recombinant lineages
  • Output structured results
    • csv
    • html?
    • fasta of all sequences that match the criteria, which enables efficient multi-pass strategies
  • filter sequences
    • by ID
    • by metadata
  • take metadata csv
  • document the output in README
  • check / fix --enabled-deletions
  • adjustable threshold for mutation prevalence
  • new color mode (with background color and monochrome text on top)
  • new bar mode (with colored lines beneath each sequence, one for each example sequence, and "intermissions" shown in the color of the "surrounding" lineage, but not as bright)
  • interactive mode, for filtering, reordering, etc.
  • sort sequences within each block
  • re-think this whole "intermission" concept
  • select a single sequence and let the tool refine the choice of parental sequences, not just focusing on commonly known lineages (going up and down in the tree)
  • use more common terms to describe things (needs feedback from people with actual experience in the field)
Owner
Lena Schimmel
Lena Schimmel
Seeing Dynamic Scene in the Dark: High-Quality Video Dataset with Mechatronic Alignment (ICCV2021)

Seeing Dynamic Scene in the Dark: High-Quality Video Dataset with Mechatronic Alignment This is a pytorch project for the paper Seeing Dynamic Scene i

DV Lab 21 Nov 28, 2022
Finding Biological Plausibility for Adversarially Robust Features via Metameric Tasks

Adversarially-Robust-Periphery Code + Data from the paper "Finding Biological Plausibility for Adversarially Robust Features via Metameric Tasks" by A

Anne Harrington 2 Feb 07, 2022
Automate issue discovery for your projects against Lightning nightly and releases.

Automated Testing for Lightning EcoSystem Projects Automate issue discovery for your projects against Lightning nightly and releases. You get CPUs, Mu

Pytorch Lightning 41 Dec 24, 2022
An implementation of RetinaNet in PyTorch.

RetinaNet An implementation of RetinaNet in PyTorch. Installation Training COCO 2017 Pascal VOC Custom Dataset Evaluation Todo Credits Installation In

Conner Vercellino 297 Jan 04, 2023
[CVPR 2022 Oral] TubeDETR: Spatio-Temporal Video Grounding with Transformers

TubeDETR: Spatio-Temporal Video Grounding with Transformers Website • STVG Demo • Paper This repository provides the code for our paper. This includes

Antoine Yang 108 Dec 27, 2022
Implementation of RegretNet with Pytorch

Dependencies are Python 3, a recent PyTorch, numpy/scipy, tqdm, future and tensorboard. Plotting with Matplotlib. Implementation of the neural network

Horris zhGu 1 Nov 05, 2021
GPU implementation of $k$-Nearest Neighbors and Shared-Nearest Neighbors

GPU implementation of kNN and SNN GPU implementation of $k$-Nearest Neighbors and Shared-Nearest Neighbors Supported by numba cuda and faiss library E

Hyeon Jeon 7 Nov 23, 2022
Pseudo-rng-app - whos needs science to make a random number when you have pseudoscience?

Pseudo-random numbers with pseudoscience rng is so complicated! Why cant we have a horoscopic, vibe-y way of calculating a random number? Why cant rng

Andrew Blance 1 Dec 27, 2021
This repository includes the official project for the paper: TransMix: Attend to Mix for Vision Transformers.

TransMix: Attend to Mix for Vision Transformers This repository includes the official project for the paper: TransMix: Attend to Mix for Vision Transf

Jie-Neng Chen 130 Jan 01, 2023
Classify bird species based on their songs using SIamese Networks and 1D dilated convolutions.

The goal is to classify different birds species based on their songs/calls. Spectrograms have been extracted from the audio samples and used as features for classification.

Aditya Dutt 9 Dec 27, 2022
SatelliteNeRF - PyTorch-based Neural Radiance Fields adapted to satellite domain

SatelliteNeRF PyTorch-based Neural Radiance Fields adapted to satellite domain.

Kai Zhang 46 Nov 20, 2022
A hybrid framework (neural mass model + ML) for SC-to-FC prediction

The current workflow simulates brain functional connectivity (FC) from structural connectivity (SC) with a neural mass model. Gradient descent is applied to optimize the parameters in the neural mass

Yilin Liu 1 Jan 26, 2022
[NeurIPS 2020] Semi-Supervision (Unlabeled Data) & Self-Supervision Improve Class-Imbalanced / Long-Tailed Learning

Rethinking the Value of Labels for Improving Class-Imbalanced Learning This repository contains the implementation code for paper: Rethinking the Valu

Yuzhe Yang 656 Dec 28, 2022
Github project for Attention-guided Temporal Coherent Video Object Matting.

Attention-guided Temporal Coherent Video Object Matting This is the Github project for our paper Attention-guided Temporal Coherent Video Object Matti

71 Dec 19, 2022
MacroTools provides a library of tools for working with Julia code and expressions.

MacroTools.jl MacroTools provides a library of tools for working with Julia code and expressions. This includes a powerful template-matching system an

FluxML 278 Dec 11, 2022
PyTorch implementation of Deformable Convolution

PyTorch implementation of Deformable Convolution !!!Warning: There is some issues in this implementation and this repo is not maintained any more, ple

Wei Ouyang 893 Dec 18, 2022
Unofficial pytorch-lightning implement of Mip-NeRF

mipnerf_pl Unofficial pytorch-lightning implement of Mip-NeRF, Here are some results generated by this repository (pre-trained models are provided bel

Jianxin Huang 159 Dec 23, 2022
Official implementation of the network presented in the paper "M4Depth: A motion-based approach for monocular depth estimation on video sequences"

M4Depth This is the reference TensorFlow implementation for training and testing depth estimation models using the method described in M4Depth: A moti

Michaël Fonder 76 Jan 03, 2023
[IROS'21] SurRoL: An Open-source Reinforcement Learning Centered and dVRK Compatible Platform for Surgical Robot Learning

SurRoL IROS 2021 SurRoL: An Open-source Reinforcement Learning Centered and dVRK Compatible Platform for Surgical Robot Learning Features dVRK compati

<a href=[email protected]"> 55 Jan 03, 2023
Deep Convolutional Generative Adversarial Networks

Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks Alec Radford, Luke Metz, Soumith Chintala All images in t

Alec Radford 3.4k Dec 29, 2022