Biology subjects

Westwood, J. T.

Publications and source records attributed to Westwood, J. T..

5 recordsLinked to original sources

bioRxiv2026

Mimic: A Unified Neural Network Architecture for Modelling Genomic Sequences and Module-level Comparison

There exist a wide range of Neural Networks (NNs) that have modelled genomic sequences well. The architectures of these NNs consist of three steps: k-mer embedding, sequence embedding, and output mapping. The NNs are primarily differentiated from one another by their sequence embedding components. Although different NNs have been compared before, these comparisons are often either limited or involve drastically different NNs. Another issue is that some NNs are configured for a specific sequence length, and different length sequences require manual modification of the architectures. This study examines the architectures of established NNs for genomics sequences. Based on these architectures, a novel NN called Mimic is developed that provides a unified architecture for the principled comparison of NN components. Mimic was used to model two different types of genomic datasets: protein-DNA interactions and enhancer activity. Along with other architecture components, different sequence embedding strategies were tried such as recurrent layers and convolutional layers. It was found that there exist specific NN components that yield significantly better performing models. Furthermore, Mimic significantly outperformed prevailing standard NNs, Bichrom and DeepSTARR, in modelling protein-DNA interactions and enhancer activity, respectively. This study provides a framework for more extensive studies toward a general prescription of NN architecture for modelling genomics data.

bioinformatics↗

bioRxiv2025

Oyster: a neural network for modelling genomic sequences that enables exact position-specific k-mer contributions

Genomic functions arise from nucleotide sequences and their overlapping k-mers - subsequences whose contributions depend on their composition, position and associations. Understanding these contributions requires computing a k-mer contribution function that may or may not consider k-mer associations. Neural networks that model associations yield powerful predictors but are notoriously hard to interpret; conversely, models that ignore associations deliver exact, position-specific contributions yet might underperform. We introduce Oyster, the first convolutional architecture that can be toggled between Exact (ignoring associations) and non-Exact modes. Exact Oysters yield closed-form k-mer contributions directly from their weights, without post-hoc attribution. By letting users choose between interpretability and complexity, Oyster provides a unified framework for transparent, high-performance sequence-to-function modeling. We apply Oyster to predict intensities of YY1-DNA interactions in human K562 cells from 500-nt DNA windows and intensities of eleven histone post-translational modifications. Exact and non-Exact variants achieved statistically indistinguishable performance, highlighting that k-mer associations are not necessarily important for all biological phenomena. Modelling of YY1-DNA interactions is dependent on the YY1 motif which is expected but is also dependent on several histone post-translational modifications including H3K9ac and H2AFZ.

genomics↗

bioRxiv2025

Novel binning-based methods for model fitting and data splitting improved machine learning imbalanced data

Machine Learning (ML) models may perform inconsistently on individual classes on nominal outputs or ranges on continuous outputs, collectively referred to here as bins. Models should be assessed through metrics that consider each bin individually, called bin metrics. Inconsistent model performance is often due to model fitting with imbalanced data. Towards improving modelling of imbalanced data, novel model fitting methods are proposed including using bin metrics as loss functions and the use of Epoch sampling. Imbalanced data also poses a challenge for appropriate data splitting. Akin split is a novel method proposed that objectively yields the most appropriate data split(s). Existing and novel model fitting methods were used to fit models, and the models were assessed by a bin metric in in two case studies. The first case study used synthetically generated datasets with different levels of noise and imbalance. On datasets with noise and greater levels of imbalance, Epoch sampling significantly improved the model performance by up to 23.6% while significantly using less resources (computation and time) by up to 57.7% compared to a standard model fitting method. The second case study used protein-genome interactions data that are often severely right-skewed. Akin split was used to split the data more appropriately than traditional methods. Model fitting methods were tried on two model configurations. The effects of the model fitting methods varied by the model configuration, but all models were significantly improved by up to 57.7% compared to the standard model fitting.

bioinformatics↗

bioRxiv2025

Deviation Error: assessing machine learning predictions for replicate measurements in genomics and beyond

Withdrawal StatementThe authors have withdrawn this manuscript because the foundational metric used throughout the study, the Deviation Error, requires substantial mathematical formalization to be established as a proper scoring rule. Completing this rigorous theoretical validation and the necessary overhaul of the accompanying synthetic case studies requires significant additional research. We intend to revisit the Deviation Error methodology in future research. Therefore, the authors do not wish this work to be cited as reference for the project. If you have any questions, please contact the corresponding author.

bioinformatics↗

bioRxiv2020

The RNA-binding protein, Rasputin/G3BP, enhances the stability and translation of its target mRNAs

G3BP RNA-binding proteins are important components of stress granules (SGs). Here we analyze the role of Drosophila G3BP, Rasputin (RIN), in unstressed cells, where RIN is not SG associated. Immunoprecipitation followed by microarray analysis identified over 550 mRNAs that copurify with RIN. The mRNAs found in SGs are long and translationally silent. In contrast, we find that RIN-bound mRNAs, which encode core components of the transcription, splicing and translation machinery, are short, stable and highly translated. We show that RIN is associated with polysomes and provide evidence for a direct role for RIN and its human homologs in stabilizing and upregulating the translation of their target mRNAs. We propose that when cells are stressed the resulting incorporation of RIN/G3BPs into SGs sequesters them away from their short target mRNAs. This would downregulate the expression of these transcripts, even though they are not incorporated into stress granules.

genomics↗