Necoda-compressed datasets.

We present Necoda, a transformative deep learning framework that compresses functional neural imaging data by over 1,000-fold while faithfully preserving signal fidelity. This work addresses the critical "data gravity" bottleneck that currently impedes research reproducibility, data sharing, and collaborative discovery in neuroscience.To demonstrate real-world utility, we applied Necoda to 5.05TB large-scale, multi-modality datasets and form a 50.1 GB data capsule. One representative case is the Allen Brain Observatory visual coding dataset. We compressed 4.82 TB of ABO dataset into a portable 4.81 GB package, along with the analysis script.