loleg / archive-forge

ORD hackathon challenge 5 https://github.com/ORD-Hackathon/hackathon-2021/issues/5

Archive-Forge

An ORD Hackathon 2021 project based on Challenge #5. This code packages data according to Frictionless Data specifications interactively in the web browser and in the console.

This is currently a prototype in three parts:

• A set of wireframes and templates for a data publishing application, in the images and templates folders, respectively.
• A frontend application based on this concept, that runs in the Web browser - see the app folder.
• A backend script which illustrates the technology and conversion and packaging process, in the root folder.

See Development setup below for deployment instructions.

Introduction

This project challenge is aimed at supporting researchers from the domain of chemical science, however our prototype is designed to be useful to the wider scientific community, making it possible to create Archive Forges for different fields of endeavour and user audiences.

Example of principle users of the Archive Forge:

• Principal investigator of a research group, who uses the tool share consistent and homogeneous data with students and faculty.
• Scientific journal editor asking authors to submit supplementary information and complementary data.

Our initial goal is to propose a web application that should make it obvious what is desired for production of a Data Package.

Based on wireframes that were elaborated during the hackathon, we have prototyped a front-end application using Vue.js and Javascript, and a back-end process using DataFlows in Python.

Wireframes of the UI are below:

**Chemistry-relevant example**

**General demo example**

Development setup

Scroll down for details on running the Python-based backend.

Frontend

Install Node.js > 12 and run, in the app folder:

npm install

Then:

npm run serve

To start a development server.

Backend

Install RDKit and Python > 3.6

To install Python libraries, run:

pip install dataflows

Alternatively, we have set up an Anaconda environment which helps to ease cross-platform issues:

1. Install Anaconda
2. Set up the archive-forge environment: conda env create -f environment.yml
3. Activate the environment: conda activate archive-forge

Once your environment is ready, you can run the data integration script:

python flow.py

An example REST server is provided for integration with the frontend, which we leave as an exercise for the reader:

python server.py

See the input and output files after running the data-integration script here below:

License

MIT License