anti-pd1-dream-examples's Introduction

Anti-PD1-DREAM-Examples

Overview

Here we describe how to build and locally run example models provided for Challenge Question 1 of the Anti-PD1 DREAM Challenge.

szabo model (R example)
tide (Python example)

Dockerizing the model

Start by cloning this repository.
Go to one of the example folders
Build the Docker image that will contain the model with the following command:
```
docker build -t awesome-antipd1-q1-model:v1 .
```

Run the model locally on synthetic data

Go to the page of the synthetic dataset provided by the Anti-PD1 DREAM challenge. This page provides useful information about the format and content of the synthetic data.
Download the file CM_026_formatted_synthetic_data_subset.tar to the location of this example folder (only available to registered participants).

Extract the content of the archive

$ tar xvf CM_026_formatted_synthetic_data_subset.tar
x CM_026_formatted_synthetic_data_subset/

Create an output folder
```
mkdir output
```

Run the dockerized model

docker run \
    -v $(pwd)/CM_026_formatted_synthetic_data_subset/:/data:ro \
    -v $(pwd)/output:/output:rw \
    awesome-antipd1-q1-model:v1

The predictions generated are saved to /output/predictions.csv.

$ cat output/predictions.csv
patientID,prediction
p267,100
p315,10
p15,4
...

Submit this model to the Anti-PD1 DREAM Challenge

This model meets the requirements for models to be submitted to Question 1 of the Anti-PD1 DREAM Challenge. Please see this page for instructions on how to submit this model.

Recommend Projects