GSoC 2023 Project Idea 10.1 Developing reanalysis tools for Allen Brain Observatory data (350 h)

DataJoint is a Houston-based company whose mission is to build and operate a science orchestration platform that eliminates barriers for researchers who rely on automation, AI, and high-performance computing. Our projects include:

  1. DataJoint Core: open-source Python and MATLAB APIs for defining and operating computational data pipelines—sequences of steps for data acquisition, processing, and transformation—via a MySQL backend.
  2. DataJoint Elements: a growing compilation of community-curated, open-source software modules for building automated data pipelines and analysis workflows for neuroscience experiments.
  3. DataJoint Works: a cloud-based platform that hosts pipelines built via Core and Elements, facilitating collaboration and reanalysis.

The Allen Institute’s Brain Observatory is a huge repository of publicly available two-photon Calcium imaging data. DataJoint’s Element Calcium Imaging offers automation in standard preprocessing methods, including Suite2p and CaImAn. This project aims to :

  1. Develop mechanism to ingest Brain Observatory data into the Element.
  2. Demonstrate reanalysis of this data via jupyter notebook
  3. Add unit tests for the above mechanism.
  4. Automate this process on the Works platform to permit replication by any Works user.

Skill level: Novice/advanced

Required skills: Python, calcium imaging data analysis

Time commitment: Half-time or full-time (350 h)

Lead mentor: Dimitri Yatsenko, Kabilar Gunalan

Project website: datajoint.com

Backup mentors: Chris Brozdowski

Tech keywords: Python, Two-Photon Imaging, Reanalysis, Replication, MySQL

Hello @dimitri @kabilar Sir!
I’m Viren Varma, I’m a senior at NIT Karnataka, I’m applying for gsoc’23. I’m relatively new to open source. But I have a vast knowledge of deep learning, computer vision and Bio-informatics through personal and course projects. I’ve got experience in working in python and data science. Both the projects from Datajoint really align with my interests and I’d love to work on it. Can you guide me on getting started with the project? Or if there’s a separate slack channel for it?
Thank you

Hi @Viren_Varma,

Thanks for your interest in our project! The first steps would be to download and explore the Allen Brain Observatory data. And to set up a development environment to run DataJoint Element Calcium Imaging. Before you begin, let’s meet virtually to discuss the project (scheduling link).

Thank you!

Kabi

Sure Sure @kabilar Sir absolutely!, I’ve sent you the meet invite. I’ll go through the link and be prepared for it.

@Viren_Varma Sounds great, thanks!

Hello @dimitri @kabilar Sir!

I’m Kshitij Soni, a senior-year student at IIPE Visakhapatnam. I am a successful GSOC 2022 contributor at ENSEMBL-EBI. I have worked as well in research experience in deep learning, computer vision and image processing, semantic segmentation, NLP, and OCR domains. I have already explored the Allen Brain Observatory Data and looking forward to getting started with this project.

Hi, it’s a bit late but I’m really interested in contributing to this project. Can you please provide next steps in writing proposal. Thanks.

Hi @kshitijsoni and @MauryaRitesh,

Thank you for your interest in our project! In order to help you plan out your proposal, let’s meet individually to discuss the project (scheduling link).

Thank you!

Kabi