PHLOWER

decomPosition of the Hodge Laplacian for inferring trajectOries from floWs of cEll diffeRentiation

Hardware requirements

PHLOWER module requires a standard computer with enough RAM to support the in-memory operations, high performance computing (HPC) cluster is recommended.

System Requirements

PHLOWER has been tested with the following OS or virtual environment:

- macOS Sonoma 14.5
- Linux 4.18.0/5.15.0
- anaconda 4.12.0

Installation

system dependencies

macOS ((install time: 15 minutes)):

• suite-sparse (>=7.8.2)
• graphviz (>=12.1.2)

  1. brew install suite-sparse
  2. brew install graphviz

  Manually install pygraphviz:
    export PATH=$(brew --prefix graphviz):$PATH
    export CFLAGS="-I $(brew --prefix graphviz)/include"
    export LDFLAGS="-L $(brew --prefix graphviz)/lib"
    pip install pygraphviz

debian (install time: 15 seconds):

• libsuitesparse-dev (>=1:5.10)
• graphviz (>=2.42.2)
• libgraphviz-dev (>=2.42.2)

  1. apt install libsuitesparse-dev
  2. apt install graphviz libgraphviz-dev

conda (install time: 1.5 minutes):

• conda-forge::suitesparse(>=5.10.1)
• graphviz (>=7.1.0)
• pygraphviz (>=1.11)

  1. conda install conda-forge::suitesparse
  2. conda install conda-forge::python-graphviz

Python dependencies

We have tested python version 3.9.0, 3.10.8, 3.10.14, 3.11.0, 3.11.5, 3.12.0.

• python (>=3.9.0)
• numpy (>=1.23.5)
• matplotlib (>=3.9.1)
• seaborn (>=0.13.2)
• networkx (>=2.8.8)
• pydot (>=1.4.2)
• igraph (>=0.10.5)
• scikit-learn (>=1.5.1)
• scipy (>=1.14.0)
• pandas (>=2.2.3)
• plotly (>=5.23.0)
• tqdm (>=4.65.0)
• leidenalg (>= 0.9.1)
• python-louvain (>=0.16)
• colorcet(>=3.0.1)
• umap-learn (>=0.5.5)
• scikit-sparse (>=0.4.8)
• scanpy (>=1.9.3)
• adjustText (>=0.8)
• pygraphviz (>=1.11)
• gudhi (>=3.10.1)
• magic-impute (>=3.0.0)
• anndata (>=0.9.2)

install from pypi (0.1.5)

Expect install time on a normal computer(Intel i5-10400 (12) @ 4.300GHz): 2 minutes

pip install phlowerpy

install newest phlower version

git clone https://github.com/CostaLab/phlower.git
cd phlower
pip install .

import phlower

Tutorial

Demo

A small scRNA-seq data Fibroblast to Neuron in 5 minutes for a computer(Intel i5-10400 (12) @ 4.300GHz).

Kidney in the paper

10X multiome data Kidney.

Regulators detection example please check Regulators.

Data info

The processed data have been deposited at: .

reproducibility

https://github.com/CostaLab/phlower-reproducibility

Reference

A preprint is available at https://doi.org/10.1101/2024.10.01.613179