Biology subjects

Kuhn, D.

Publications and source records attributed to Kuhn, D..

4 recordsLinked to original sources

bioRxiv2026

Model Validation Protocols for Machine Learning in Small Molecule Drug Discovery

Machine learning (ML) models for molecular property prediction are increasingly deployed in drug discovery, yet their adoption in real-world scenarios requires an understanding of the conditions in which a model succeeds or fails. While standardized benchmarks are powerful instruments to measure and unlock progress in ML research, they should not be blindly treated as the end goal. Especially static and retrospective benchmarks, in which no true unknown test set is employed, limit our ability to robustly validate a model's performance. Building on the collective expertise of a cross-industry consortium, we present a model validation framework consisting of five recommendations that would enable the community to move beyond aggregate metrics toward understanding where and why molecular property prediction models fail. We connect evaluation choices to real-world applications and case studies encountered in pharmaceutical research. The framework proposes splitting strategies that mimic realistic distribution shifts and expose common failure modes. We apply the recommended framework to a recently released dataset of absorption, distribution, metabolism, and excretion (ADME) properties. Across two complementary model algorithms, our case studies reveal four distinct failure modes (extrapolation, interpolation, representation, and evaluation), showing that model errors arise not only from distribution shift but also from limitations in molecular representations. Our results show that commonly used evaluation protocols can significantly overestimate performance and may not detect important model failure modes. All software and data are released via https://github.com/srijitseal/polaris.

bioinformatics↗

bioRxiv2022

SUMO - In Silico Sequence Assessment Using Multiple Optimization Parameters

To select the most promising screening hits from antibody and VHH display campaigns for subsequent in-depth profiling and optimization, it is highly desirable to assess and select sequences on properties beyond only their binding signals from the sorting process. In addition, developability risk criteria, sequence diversity and the anticipated complexity for sequence optimization are relevant attributes for hit selection and optimization. Here, we describe an approach for the in silico developability assessment of antibody and VHH sequences. This method not only allows for ranking and filtering multiple sequences with regard to their predicted developability properties and diversity, but also visualizes relevant sequence and structural features of potentially problematic regions and thereby provides rationales and starting points for multi-parameter sequence optimization.

bioinformatics↗

bioRxiv2022

Three million images and morphological profiles of cells treated with matched chemical and genetic perturbations

Identifying genetic and chemical perturbations with similar impacts on cell morphology can reveal compounds mechanisms of action or novel regulators of genetic pathways. Research on methods for identifying such similarities has lagged due to a lack of carefully designed and well-annotated image sets of cells treated with chemical and genetic perturbations. Here, we create such a Resource dataset, CPJUMP1, where each perturbed gene is a known target of at least two chemical compounds in the dataset. We systematically explore the directionality of correlations among perturbations that target the same gene, and we find that identifying matches between chemical perturbations and genetic perturbations is a challenging task. Our dataset and baseline analyses provide a benchmark for evaluating methods that measure perturbation similarities and impact, and more generally, learn effective representations of cellular state from microscopy images. Such advancements would accelerate the applications of image-based profiling, such as functional genomics and drug discovery.

bioinformatics↗

bioRxiv2019

The Avocado Genome Informs Deep Angiosperm Phylogeny, Highlights Introgressive Hybridization, and Reveals Pathogen Influenced Gene Space Adaptation

The avocado, Persea americana, is a fruit crop of immense importance to Mexican agriculture with an increasing demand worldwide. Avocado lies in the anciently-diverged magnoliid clade of angiosperms, which has a controversial phylogenetic position relative to eudicots and monocots. We sequenced the nuclear genomes of the Mexican avocado race, P. americana var. drymifolia, and the most commercially popular hybrid cultivar, Hass, and anchored the latter to chromosomes using a genetic map. Resequencing of Guatemalan and West Indian varieties revealed that [~]39% of the Hass genome represents Guatemalan source regions introgressed into a Mexican race background. Some introgressed blocks are extremely large, consistent with the recent origin of the cultivar. The avocado lineage experienced two lineage-specific polyploidy events during its evolutionary history. Although gene-tree/species-tree phylogenomic results are inconclusive, syntenic ortholog distances to other species place avocado as sister to the enormous monocot and eudicot lineages combined. Duplicate genes descending from polyploidy augmented the transcription factor diversity of avocado, while tandem duplicates enhanced the secondary metabolism of the species. Phenylpropanoid biosynthesis, known to be elicited by Colletotrichum (anthracnose) pathogen infection in avocado, is one enriched function among tandems. Furthermore, transcriptome data show that tandem duplicates are significantly up- and down-regulated in response to anthracnose infection, whereas polyploid duplicates are not, supporting the general view that collections of tandem duplicates contribute evolutionarily recent \"tuning knobs\" in the genome adaptive landscapes of given species.\n\nSIGNIFICANCE STATEMENTAvocado is a nutritious, economically important fruit species that occupies an unresolved position near the earliest evolutionary branchings of flowering plants. Our nuclear genome sequences of Mexican and Hass variety avocados inform ancient evolutionary relationships and genome doublings, the admixed nature of Hass, and provide a look at how pathogen interactions have shaped avocados more recent genomic evolutionary history.

genomics↗