/MetaLogo

MetaLogo: a heterogeneity-aware sequence logo generator and aligner

Primary LanguagePythonApache License 2.0Apache-2.0

MetaLogo

Introduction

MetaLogo is a tool for making sequence logos. It can take multiple sequences as input, automatically identify the homogeneity and heterogeneity among sequences and cluster them into different groups given any wanted resolution, finally output multiple aligned sequence logos in one figure. Grouping can also be specified by users, such as grouping by lengths, grouping by sample Id, etc. Compared to conventional sequence logo generator, MetaLogo can display the total sequence population in a more detailed, dynamic and informative view.

To use MetaLogo, you could visit our public webserver http://metalogo.omicsnet.org. You could also install MetaLogo as a python package to using MetaLogo in your python scripts or in your OS terminal. If you want to provide MetaLogo to people in your local network, you could also setup a webserver by using docker.

Please check the tutorial for detailed usage of MetaLogo package and webserver.

Introduction

Install

Before you install MetaLogo, you could first set up a virtual environment for MetaLogo using conda:

$conda create -n metalogo python=3.7
$conda activate metalogo

Then, you can install MetaLogo using following commands:

$git clone https://github.com/labomics/MetaLogo 
$cd MetaLogo
$pip install .

If succeed, you can run MetaLogo directly:

$metalogo -h

Note that the paths of clustal omega and fasttree bins should be given to MetaLogo, which are required for grouping. You can have these binary files in the dependencies folder in this repo. Basically, you can run metalogo as follows:

$metalogo --seq_file examples/cdr3.fa --clustalo_bin dependencies/clustalo --fasttree_bin dependencies/FastTree --fasttreemp_bin dependencies/FastTreeMP --logo_format pdf --output_name test.xx --task_name test.xx

If you want to using MetaLogo in your scripts, it is also very easy:

from MetaLogo import logo
sequences = [['seq1','ATACAGATACACATCACAG'],['seq2','ATACAGAGATACCAACAGAC'],['seq3','ATACAGAGTTACCCACGGAC']]

bin_args = {
    'clustalo_bin':'../MetaLogo/dependencies/clustalo',
    'fasttree_bin':'../MetaLogo/dependencies/FastTree',
    'fasttreemp_bin':'../MetaLogo/dependencies/FastTreeMP',
    }

lg = logo.LogoGroup(sequences,height_algorithm='probabilities',group_strategy='length', **bin_args)
lg.draw()

Note that the paths of clustal omega and fasttree bins should be given to MetaLogo, which are required for grouping.

MetaLogo also provide webserver to draw sequence logos without coding. For server setup, you need to install the required packages first.

$pip install .[webserver]

You need to start a redis server and the redis queue servic before running the MetaLogo server locally (see tutorial). We recommend that you can build a docker container to provide the service:

$cat MetaLogo/server.docker.sh
 docker build -t metalogo:v1 .
 docker run -d  --expose 8050 --name metalogo -e VIRTUAL_HOST=metalogo.omicsnet.org -v "$(pwd)":/code metalogo:v1 
$sh MetaLogo/server.docker.sh
 ...
$docker ps
 CONTAINER ID    IMAGE          COMMAND                  CREATED      STATUS      PORTS       NAMES
 ad598ca936df    metalogo:v1    "/bin/sh -c 'GUNICORN"   2 days ago   Up 2 days   8050/tcp    metalogo

Docker needs to be installed in the system before running the command. This command will build the Docker image and start a Docker container. You could set a nginx-proxy layer to redirect network requests to MetaLogo container or you can just simply use the MetaLogo docker to receive outside network traffic from your local network.

More detailed instructions please check the tutorial of MetaLogo.

Citation

Please kindly cite our work if you use MetaLogo in your work.

Chen, Yaowen, et al. "MetaLogo: a heterogeneity-aware sequence logo generator and aligner." Briefings in Bioinformatics (2021).