marakeby / pnet_prostate_paper Goto Github PK

P-NET, Biologically informed deep neural network for prostate cancer classification and discovery

License: GNU General Public License v2.0

Python 12.12% Jupyter Notebook 35.84% HTML 51.95% R 0.09%

pnet_prostate_paper's Introduction

P-NET

P-NET, Biologically informed deep neural network for prostate cancer classification and discovery

view interactive network architecture · ·

About The Project
Getting Started
- Prerequisites
- Installation
Usage
References
License
Contact
Acknowledgements

About The Project

Biologically informed deep neural network for prostate cancer classification and discovery

Getting Started

To get a local copy up and running, follow these simple steps

Prerequisites

python 2.7, check environments.yml for list of needed packages

Installation

Clone the repo

git clone https://github.com/marakeby/pnet_prostate_paper.git

Create conda environment

conda env create --name pnet_env --file=environment.yml

Based on your use, you may need to download one or more of the following

a. Data files (needed to retrain models and generate figures). Extract the files under _database directory. If you like to store it somewhere else, you may need to set the DATA_PATH variable in config_path.py accordingly.

b. Log files (needed to regenerate paper figures). Extract the files under _logs directory. If you like to store it somewhere else, you may need to set the LOG_PATH variable in config_path.py accordingly.

c. Plots files (a copy of the paper images). Extract the files under _plots directory. If you like to store it somewhere else, you may need to set the PLOTS_PATH variable in config_path.py accordingly.

Usage

Activate the created conda environment
```
source activate pnet_env
```

Add the current diretory to PYTHONPATH, e.g.

export PYTHONPATH=~/pnet_prostate_paper:$PYTHONPATH

To generate all paper figures, run
```
cd ./analysis
python run_it_all.py
```
To generate individual paper figure run the different files under the 'analysis' directory, e.g.
```
cd ./analysis
python figure_1_d_auc_prc.py
```
For Figure3 , make sure you run prepare_data.py before running other files
To re-train a model from scratch run
```
cd ./train
python run_me.py
```
This will run an experiment 'pnet/onsplit_average_reg_10_tanh_large_testing' which trains a P-NET model on a training-testing data split of Armenia et al data set and compare it to a simple logistic regression model. The results of the experiment will be stored under _logsin a directory with the same name as the experiment.
To run another experiment, you may uncomment one of the lines in the run_me.py to run the corresponding experiment. Note that some models especially cross validation experiments may be time consuming.

License

Distributed under the GPL-2.0 License License. See LICENSE for more information.

Contact

Haitham - @HMarakeby

Project Link: https://github.com/marakeby/pnet_prostate_paper

References

Elmarakeby H, et al. "Biologically informed deep neural network for prostate cancer classification and discovery." Nature. Online September 22, 2021. DOI: 10.1038/s41586-021-03922-4
Armenia, Joshua, et al. "The long tail of oncogenic drivers in prostate cancer." Nature genetics 50.5 (2018): 645-651.
Robinson, Dan R., et al. "Integrative clinical genomics of metastatic cancer." Nature 548.7667 (2017): 297-303.
Fraser, Michael, et al. "Genomic hallmarks of localized, non-indolent prostate cancer." Nature 541.7637 (2017): 359-364.

Acknowledgements

This work was supported in part by the Fund for Innovation in Cancer Informatics, Mark Foundation, Prostate Cancer Foundation, Movember, and the National Cancer Institute at the National Institutes of Health.

pnet_prostate_paper's People

Contributors

Stargazers

Watchers

Forkers

ahmedarslan jianguozhou3 liujilei156231 qasemabdullah exascale-genomics smallcraft dudzicp xl60-hust shoujiang dolittle007 andrewsris belaad yelongb wukaiyeah bit-vs-it terragord7 shunsunsun allensmile stjordanis 5l1v3r1 mglnbk yiluheihei manishkth wangbostc zheyanliu99 mubashermohammed xiangbu zj-zhang sudolin yisnsiy collvey yutingdai1125 guliba catherinehui petaleaf shicheng-guo yu-zhang-oyo amitbioinfo causalai mborhi iff-0303 caozhanhang kancheng explcre wangdi2016 harel-coffee qinqian hosseini7798 wangze09 1830416002 hyeongd ychienhung yanluocityu

pnet_prostate_paper's Issues

A request to latest code.

I found that your code has been updated recently. python2 and python3 versions, tensorflow1.x and tensorflow2.x are both used in your code.I cannot run your code. Can you give me the latest code that can be run directly? Thank you very much for your kind consideration.

Error: No packages found in current win-32 channels matching: imageio

I am trying on windows conda env create --name pnet_env --file=environment.yml and getting this error

Using Anaconda Cloud api site https://api.anaconda.org
Fetching package metadata: ....
Error: No packages found in current win-32 channels matching: imageio

when I am using same versions as mentioned in yml file, it is giving error of package version not found error.
conda env create --name pnet_env --file=environment.yml
Using Anaconda Cloud api site https://api.anaconda.org
Using Anaconda Cloud api site https://api.anaconda.org
Fetching package metadata: ..........
Error: No packages found in current win-32 channels matching: pip 18.1*

No such file or directory pathways_short_names.xlsx'

I tried to regenerate figure3 in the paper,but there is no 'pathways_short_names.xlsx' file under '~/_database/pathways' directory.
Could you please supply this file?
I am looking forward to your reply!!!
Thank you for your time.

environment problem

Problem 1, the lifelines in environment.yml need to install the dependency package Bottleneck, pip will install Bottleneck1.3.4 by default, but there are problems causing Bottleneck1.3.4 will always fail to install, so need to install Bottleneck1.3.2 in advance.

Problem 2, h5py in environment.yml did not find version 2.9.01.c, so install 2.9.0

Does PNET support training with multiple classes?

Dear @marakeby,

Thanks for developing and releasing PNET. I am trying to do exploratory analyses in some multimodal data using this framework. Labels are more than two, so I wonder whether a multiclass flag is already available by default in one of the reported architectures tested in the paper. I think small changes in the nn.py and/or model_params_ dictionary could allow this, yet I have been testing without much success (e.g. 'prediction_outputto None, and a alternativeloss` key).

Do you think there are already defaults for trying this? Thanks for any comments on this.

Extended figures

Where is the code for extended figures 6 and 8?
Thanks

how to incorporate new loss function (your own) in PNET architecture

Any suggestion or comments with reference to PNET architecture will be helpful.

"ResolvePackageNotFound: - h5py=2.9.0l.c" Error

Hi,
I tried to create a p-net environment with "conda env create --name pnet_env --file=environment.yml", but failed. The messages are:

Collecting package metadata (repodata.json): done
Solving environment: failed

ResolvePackageNotFound: 
  - h5py=2.9.0l.c

So, how to solve this problem? Thanks.

No such file : pathways_short_names.xlsx

environment.yml ERROR

File configuration error
"""
dependencies:
...
h5py=2.9.0l.c
...
"""
should be modify.
I try to change it:
"""
dependencies:
...
h5py
...
"""
the code is working.
However, I am not sure whether it will affect the subsequent analysis.
We look forward to your reply or solution !!

Thank you for your time

Environment fails at installing decorator: not compatible with Python 2.7

During installation I found a conflict with the decorator package version that is installed. Namely, the default version conda installs gets into conflict because it gets in conflict with python 2.7 , a default requirement.

I think this could be solved by downgrading decorator to a 2.6+ compatible version, I found that decorator 4.3.2 is valid based on the current decorator release list.

pip install decorator==4.3.2

Could you update this dependency into the environment.yml file? It could probably ease the installation process.

Thanks for dev and sharing of this repository.

Environment Error

Hi, Can someone please tell me how to rectify the environment error. I am encountering this error for a few days now and the code does not run. My base python version in 3.9. I am also attaching screenshots for the same. Any leads/help would be appreciated.

Thanks