Neural signal propagation atlas [1], genome [2], and single-cell transcriptome [3], neuropeptide/GPCR deorphanization [4], anatomical connectome [5,6], monoaminergic connectome [7], and chemical-synapse sign predictions [8] all in one place. Worm Neuro Atlas allows to build a basic version of the neuropeptidergic connectome [9] ([9] contains more detailed analysis). Please cite the code if you use it, along with the papers containing the datasets you use.
Read the full documentation here https://francescorandi.github.io/wormneuroatlas/
Run a demo of Worm Neuro Atlas in a Jupyter notebook on Colab here https://colab.research.google.com/drive/1j59Mv_PSaD4Nj2ITNDQWVxWfrl6ssqXy?usp=sharing
wormenuroatlas can be installed via pip
pip install wormneuroatlas
# or
python -m pip install wormneuroatlas
or by cloning this repository and pip install the module, e.g. via
git clone git@github.com:francescorandi/wormneuroatlas.git
cd wormneuroatlas
python -m pip install .
NeuroAtlas is the main class that aggregates all the datasets, and directly handles the Signal Propagation Atlas, the anatomical connectome, and the monoaminergic connectome [7].
You can access the wild-type and unc-31 signal propagation atlas, for example, via
NeuroAtlas.get_signal_propagation_atlas(strain="wt") # or strain="unc31"
NeuroAtlas will use the other classes of the Python module (WormFunctionalConnectivity, WormBase, Cengen, PeptideGPCR, ...) to access data from other datasets like neural signal propagation, genome, single-cell transcriptome and aggregate those datasets. For example, you can get gene expression levels of genes flp-1 and aqp-1 in neurons AVAL and AVDR via
NeuroAtlas.get_gene_expression(gene_names=["flp-1","aqp-1"], neuron_names=["AVAL","AVDR"]).
See Examples for more information on how to access the various datasets.
NeuroAtlas also manages the conversions between different conventions for neural IDs. NeuroAtlas can be instantiated to maintain the "exact" neural identities, or to merge neurons into classes (i.e. to approximate neuron identities): merge_bilateral=True will merge results for, e.g., AVAL and AVAR into the class AVA_, merge_dorsoventral=True will merge RMED and RMEV into RME_, while merge_numbered=True will merge VB3, VB4, ... into VB. These options can be combined to merge, for example, SMBVL, SMBVR, SMBDL, and SMBDR into SMB__ with merge_bilateral=True, merge_dorsoventral=True.
The Cengen class interfaces with single-cell RNASeq database from the CeNGEN project. Its main function is Cengen.get_gene_expression(), which is called by NeuroAtlas.get_gene_expression() after converting neuron IDs to CeNGEN-style IDs.
The PeptideGPCR class provides an interface to the neuropeptide/GPCR deorphanization in [4]. The two main functions are
PeptideGPCR.get_gpcrs_binding_to(peptides)
and
PeptideGPCR.get_peptides_binding_to(gpcrs)
which return the GPCRs binding to given peptides and the peptides binding to given GPCRs, respectively.
The WormBase class uses the REST API provided by wormbase.org. WormBase currently has methods to retrieve lists of transcripts for given genes, as well as functions to convert WormBase-style gene IDs to gene names, etc.
- plot_signal_propagation.py shows you how to access the signal propagation data,
- gene_expression.py shows you how to get gene expression data from CeNGEN,
- peptideGPCR_binding.py shows you how to get the peptides binding to given GPCRs and vice versa,
- make_peptidergic_connectome.py shows you how to combine these functions to compile the neuropeptidergic connectome [9] using gene expression and neuropeptide/GPCR deorphanization.
- Randi et al., arXiv 2022 https://arxiv.org/abs/2208.04790
- WormBase, wormbase.org
- Taylor et al., Cell 2021
- Beets et al., bioRxiv 2022 https://www.biorxiv.org/content/10.1101/2022.10.30.514428v1
- White et al., Phil. Trans. R. Soc 1986
- Witvliet et al., Nature 2021
- Bentley et al., PLOS Comp. Bio. 2016
- Fenyves et al., PLOS Comp. Bio. 2020
- Ripoll-Sanchez et al., bioRxiv 2022 https://www.biorxiv.org/content/10.1101/2022.10.30.514396v2