Mario Stanke

Extended Tiberius models outscore Helixer and ANNEVO on all 33 genomes from the five clades all three tools cover, and approach BRAKER3 accuracy in four clades without extrinsic evidence.

Vipsania is an unsupervised deep learning model for gene finding in eukaryotic genomes, achieving higher accuracy than supervised methods across most clades.

The paper annotates 200 insect genomes using BRAKER3 and VARUS to enable consistent comparative genomics analyses across species.

Tiberius is a deep learning gene predictor that integrates HMMs with CNNs and LSTMs, achieving high accuracy in mammalian genomes.

BRAKER3 improves eukaryotic genome annotation by integrating RNA-seq, protein data, and statistical models, achieving higher accuracy than previous tools.

GALBA is a fully automated gene annotation pipeline combining miniprot and AUGUSTUS, achieving high accuracy in large vertebrate genomes and new species.

The paper introduces geometric hashing, a method to filter alignment seeds with high specificity and efficiency, reducing false positives in homology searches across human and mouse genomes.

ClaMSA uses a novel CTMC layer to classify coding regions in genome alignments, achieving six times fewer false positives than existing methods on vertebrate and fly data.

The paper generates a chromosome-level genome assembly for the functionally extinct northern white rhinoceros and compares it to the southern white rhinoceros.

TSEBRA selects gene predictions from BRAKER1 and BRAKER2 outputs by comparing transcript scores based on RNA-seq and protein evidence, achieving higher accuracy than either pipeline alone.

BRAKER2 improves eukaryotic genome annotation by integrating GeneMark-EP+ and AUGUSTUS with cross-species protein sequences, achieving higher accuracy than MAKER2 and BRAKER1 in certain conditions.

VARUS samples RNA reads from the Sequence Read Archive to enable automated genome annotation with high sensitivity and specificity.

The paper introduces CAT, a toolkit for simultaneous clade and personal genome annotation, demonstrating its ability to discover novel genes and improve cross-species RNA expression analysis.

The paper compares genome assemblies of Mus caroli and Mus pahari with Mus musculus and Rattus norvegicus to reveal lineage-specific evolutionary mechanisms, particularly repeat-associated processes.

The paper presents full-length de novo genome assemblies for 16 inbred mouse strains, identifying strain-specific haplotypes and novel functional loci, including a large unannotated gene with potential developmental roles.

The paper analyzes the house spider genome to identify an ancient whole-genome duplication shared by spiders and scorpions, distinct from horseshoe crab duplications.

This study sequenced wild tobacco genomes to uncover how nicotine biosynthesis evolved through gene duplications and transposable elements.