SHERLOCK: A Python pipeline that transforms raw space telescope data into exoplanet discoveries, hunting for hidden worlds in TESS and Kepler observations that automated systems missed.

Every 27 days, NASA’s TESS spacecraft completes another sweep of the cosmos, capturing the subtle dimming of starlight that betrays the presence of orbiting worlds. But in the torrents of photometric data streaming back to Earth, countless planetary candidates slip through the cracks of automated detection algorithms—victims of detection thresholds, poor data quality, or simply the overwhelming scale of the search.

SHERLOCK transforms this challenge into opportunity with a comprehensive six-module pipeline that democratizes exoplanet discovery. Starting with raw light curves from TESS and Kepler archives, it systematically searches for transit signatures, vets potential candidates against false positives, performs statistical validation, refines orbital parameters through sophisticated modeling, and even calculates optimal ground-based follow-up windows. The entire workflow is orchestrated through a simple YAML configuration file, making professional-grade planet hunting accessible to researchers without requiring deep pipeline expertise.

Already powering cutting-edge science, SHERLOCK drives the SPECULOOS survey’s hunt for Earth-sized worlds in the habitable zones of ultra-cool dwarfs—prime targets for atmospheric characterization with next-generation telescopes. It’s also enabling the FATE project’s unprecedented search for planets orbiting evolved hot subdwarfs, probing how planetary systems survive their host star’s violent post-main-sequence evolution. For astronomers ready to join the exoplanet frontier, SHERLOCK provides the computational firepower to turn space-based observations into world-changing discoveries.

⭐ Stars: 6
💻 Language: Jupyter Notebook
🔗 Repository: iaa-so-training/sherlock-tutorial