Metadata Extraction and JSON Export Pipeline for LAGE Sequencing Data

Presentation slides from the MDMC thesis defence on a modular Python pipeline for automated metadata extraction and FAIR-compliant packaging of high-throughput sequencing data at LAGE (Laboratory of Applied Genomics and Epigenomics), Area Science Park, Trieste. The pipeline addresses three problems in sequencing data curation: metadata scattered across three instrument platforms (Illumina NovaSeq 6000, Oxford Nanopore PromethION 24, Illumina iScan), the absence of cross-platform sample lineage, and the lack of machine-readable FAIR output. It uses content-based detection to identify files regardless of name or location, extracts metadata into a common JSON schema, reconstructs sample provenance across runs and platforms, and packages each dataset as an RO-Crate 1.2 research object. These slides cover the motivation, pipeline architecture, the modular extractor design, worked examples answering real curation questions, FAIR compliance through RO-Crate, and directions for future work.