Sebastian Böcker

Chemistry-based vectors map the chemical space of natural biomes from untargeted mass spectrometry data

The paper uses chemistry-based vectors to analyze untargeted mass spectrometry data and identifies compound classes that differentiate natural biomes.

Small molecule machine learning: All models are wrong, some may not even be useful

The paper evaluates coverage bias in small molecule datasets and proposes methods to assess and improve dataset representativeness for machine learning models.

MAD HATTER correctly annotates 98% of small molecule tandem mass spectra searching in PubChem

The paper introduces a novel method that achieves 97.6% correct annotation of small molecule tandem mass spectra by combining spectral data with structural database information.

BERTax: taxonomic classification of DNA sequences with Deep Neural Networks

BERTax uses deep learning to classify DNA sequences taxonomically without requiring known database relatives, outperforming existing methods for novel organisms and combining with databases to improve accuracy.

MSNovelist: De novo structure generation from mass spectra

MSNovelist generates small molecule structures de novo from mass spectra, outperforming database-based methods for novel compounds.

Assigning confidence to structural annotations from mass spectra with COSMIC

COSMIC assigns confidence to structural annotations from mass spectra by combining database generation, in silico annotation, and a confidence score, outperforming existing methods and annotating novel structures.

Ion Identity Molecular Networking in the GNPS Environment

The paper introduces Ion Identity Molecular Networking to connect ion species of the same molecule, improving network connectivity and annotation.

Classes for the masses: Systematic classification of unknowns using fragmentation spectra

CANOPUS is a tool that predicts compound classes from fragmentation spectra, enabling annotation of unknown compounds without reference data.

Chemically-informed Analyses of Metabolomics Mass Spectrometry Data with Qemistree

The paper introduces Qemistree, a method for analyzing metabolomics mass spectrometry data using hierarchical molecular fingerprints and chemical ontologies to explore chemical composition.

ZODIAC: database-independent molecular formula annotation using Gibbs sampling reveals unknown small molecules

ZODIAC improves molecular formula annotation in metabolomics by combining SIRIUS results with Gibbs sampling, reducing errors and discovering novel compounds.

Feature-based Molecular Networking in the GNPS Analysis Environment

The paper introduces FBMN, a method for improving molecular networking by enhancing quantitative analysis and isomer distinction using feature detection and alignment tools in the GNPS platform.

Significance estimation for large scale untargeted metabolomics annotations

The paper develops methods to estimate false discovery rates in untargeted metabolomics annotations and shows how adjusting spectral matching parameters can significantly increase the number of annotations.