Multi‑tier pharmacovigilance analysis of FAERS for six ultra‑rare neurological diseases: code and processed data

This repository contains the complete Python analysis pipeline (Colab notebook) and processed summary tables used in the pharmacovigilance study of FAERS data for Batten disease (CLN2), PKAN, Niemann‑Pick type C, Sanfilippo syndrome, GM2 gangliosidosis, and Friedreich's ataxia. The pipeline performs disproportionality analysis, FDA label comparison, PubMed novelty scoring, polypharmacy adjustment, and sensitivity analyses. All outputs include the 25 novel unlabelled drug‑related safety signals, 7 monotherapy‑robust signals, E‑values, negative control analysis, MedDRA SOC mapping, and publication‑ready figures