Biology subjects

Van Kleunen, L. B.

Publications and source records attributed to Van Kleunen, L. B..

5 recordsLinked to original sources

bioRxiv2026

Cross-assay RNA modeling reveals cancer biomarkers

The clinical heterogeneity of cancer poses a major challenge for precision medicine. Limited cohort sizes across evolving assay platforms impede reliable biomarker discovery. Here, we systematically evaluate how to integrate data from four transcriptomics platforms: bulk and single-cell (sc) RNA sequencing (RNA-seq), NanoString, and microarray for predictive modeling in cancer. We use high-grade serous carcinoma (HGSC) of tube-ovarian origin as a model system, as it is highly heterogeneous in both biology and assay data. We find that using fold-change of gene expression in patients with matched pre- and post-neoadjuvant chemotherapy samples reduces inter-patient and inter-assay variability but is insufficient to overcome platform-specific biases. Microarray and scRNA-seq data exhibit systematic biases, while RNA-seq and NanoString show the most promise for combination into a single training cohort. To mitigate inter-assay limitations, we generate a new data set of HGSC tumor samples profiled with both RNA-seq and NanoString, and use it to identify the limits of detection and optimal harmonization strategies. Our approaches enable integration of cohorts for separate and combined RNA-seq and NanoString predictive models of disease recurrence (test-set AUROCs > 0.8), validated in external microarray cohorts. We leverage single-cell and bulk RNA-seq network-based analyses to provide mechanistic context for genes in the predictive models. Our models indicate that GBP4 expression is a key predictor of recurrence and marks immune remodeling towards cytotoxicity. We provide an interactive web portal to facilitate exploration of data and results. These findings guide cross-assay harmonization of transcriptomic data and enable improved predictive modeling in heterogeneous cancers. Statement of SignificanceWe present a framework for integrating RNA-seq, NanoString, microarray, and single-cell transcriptomic data for predictive modeling, enabling robust biomarker discovery in heterogeneous cancers and identifying GBP4 as a marker of immune remodeling.

bioinformatics↗

bioRxiv2026

On the predictability of progression-free survival in ovarian cancer from NanoString gene expression data

In the treatment of high grade serous ovarian cancer (HGSC), patients initially diagnosed with unresectable tumors are first treated with neoadjuvant chemotherapy (NACT) to reduce tumor burden prior to surgery. Analysis of matched pre- and post-NACT samples from the same patients enables the investigation of chemotherapy impacts and the biomarkers of progression. Although the tumor immune microenvironment (TIME) has increasingly been recognized as critical in shaping the development and progression of HGSC, we lack a comprehensive understanding of how chemotherapy remodels the TIME. Previous studies have found evidence for a general inflammatory response post-NACT, despite inconsistencies regarding which differentially expressed genes and pathways are implicated. We combine matched NanoString gene expression data from multiple sources to create a large dataset of matched pre- and post- NACT samples (N=83, with 29 novel to this study) and investigate reproducibility. Further, we use machine learning methods to investigate whether patient progression-free survival (PFS) can be predicted from the observed impact of chemotherapy on the TIME as represented by the comprehensive set of NanoString features. We find overall low predictability of PFS from all NanoString features, suggesting that previous results may have been limited by small sample size effects and that larger datasets are needed to identify more generalizable and translatable findings. We identify a set of differential expression features that are the most important for predicting patient outcomes that can be validated in future computational and biological studies. Author summaryA subset of patients with high grade serous ovarian cancer are treated with chemotherapy before surgery to reduce tumor burden. We investigate a large dataset of samples taken before and after chemotherapy. These matched samples enable an investigation of how the environment around tumors, for example immune cell infiltration, reacts to chemotherapy, providing insights into biomarkers for treatment response and treatments that could complement chemotherapy. This larger dataset only partially replicates results from previous studies, while also providing new insights. Machine learning models designed to predict the time to patient recurrence from available biomarkers indicate that they are not strongly predictive of patient outcomes, in contrast to past studies. These results suggest that larger datasets are needed. We identify a set of genes that change with chemotherapy and are indicative of and potentially useful for predicting time to disease recurrence and can be further investigated.

cancer biology↗

bioRxiv2026

Understanding and addressing under-sampling in plant-pollinator networks using stacked models for missing link prediction

Pollinator diversity and pollination are under threat from anthropogenic disturbances. Plant-pollinator networks are useful tools to study the consequences of these disturbances, but their links are often under-sampled, potentially biasing conclusions. We introduce a computational method that can be used to predict which unobserved links in a plant-pollinator network dataset are missing. This method extends a state-of-the-art machine learning approach to missing link prediction based on stacked generalization, which has been promising in other contexts, to the setting of bipartite plant-pollinator networks with species traits treated as node attributes. This approach predicts missing links using an ensemble of predictors based on both observed network topology and species traits. We demonstrate this approach on synthetic and empirical plant-pollinator networks. We take advantage of a unique empirical plant-pollinator dataset which samples from three sites using (a) records of observed visits of pollinators to plant species and (b) pollen analysis with detailed species trait annotations for both plant and pollinator nodes. Visit-based sampling under-samples the links in these networks, whereas pollen-based sampling identifies additional links, allowing us to investigate missing link prediction from realistic patterns of field under-sampling. We show that full plant-pollinator networks can be partially reconstructed from the under-sampled networks using our approach. Further, we show that this method can be used to identify which features are the most important for predicting missing links, which allows for investigation into mechanisms driving plant-pollinator link formation and observation. This method is broadly applicable to under-sampled plant-pollinator networks.

ecology↗

bioRxiv2024

Predicting missing links in food webs using stacked models and species traits

Networks are a powerful way to represent the complexity of complex ecological systems. However, most ecological networks are incompletely observed, e.g., food webs typically contain only partial lists of species interactions. Computational methods for inferring such missing links from observed networks can facilitate field work and investigations of the ecological processes that shape food webs. Here, we describe a stacked generalization approach to predicting missing links in food webs that can learn to optimally combine both structural and trait-based predictions, while accounting for link direction and ecological assumptions. Tests of this method on synthetic food webs show that it performs very well on networks with strong group structure, strong trait structure, and various combinations thereof. Applied to a global database of 290 food webs, the method often achieves near-perfect performance for missing link prediction, and performs better when it can exploit both species traits and patterns in connectivity. Furthermore, we find that link predictability varies with ecosystem type, correlates with certain network characteristics like size, and is principally driven by a subset of ecologically-interpretable predictors. These results indicate broad applicability of stacked generalization for studying ecological interactions and understanding the processes that drive link formation in food webs.

ecology↗

bioRxiv2024

The spatial structure of the tumor immune microenvironment can explain and predict patient response in high-grade serous carcinoma

Despite ovarian cancer being the deadliest gynecological malignancy, there has been little change to therapeutic options and mortality rates over the last three decades. Recent studies indicate that the composition of the tumor immune microenvironment (TIME) influences patient outcomes but are limited by a lack of spatial understanding. We performed multiplexed ion beam imaging (MIBI) on 83 human high-grade serous carcinoma tumors -- one of the largest protein-based, spatially-intact, single-cell resolution tumor datasets assembled -- and used statistical and machine learning approaches to connect features of the TIME spatial organization to patient outcomes. Along with traditional clinical/immunohistochemical attributes and indicators of TIME composition, we found that several features of TIME spatial organization had significant univariate correlations and/or high relative importance in high-dimensional predictive models. The top performing predictive model for patient progression-free survival (PFS) used a combination of TIME composition and spatial features. Results demonstrate the importance of spatial structure in understanding how the TIME contributes to treatment outcomes. Furthermore, the present study provides a generalizable roadmap for spatial analyses of the TIME in ovarian cancer research.

cancer biology↗