Biology subjects

Dee, W.

Publications and source records attributed to Dee, W..

6 recordsLinked to original sources

bioRxiv2026

SHOT-CCR: Biologically guided adversarial training for test-time adaptation in cellular morphology

AO_SCPLOWBSTRACTC_SCPLOWPervasive batch effects are a common issue, especially in recent large-scale Cell Painting datasets, which have been produced to aid AI-enhanced drug discovery efforts. Technical differences arising from experiments carried out in different batches can cause models to fail to generalize to unseen batches, despite good predictive performance "within batch". We propose a biologically grounded test-time adaptation framework, SHOT-CCR, which uses cell-invariant gradient reversal to decouple morphological signal from experimental confounders. Our approach performs 4.5% better than the current RxRx1 benchmark, classifying 1,139 classes of siRNA genetic perturbations with 91.6% accuracy. We deliver consistent results over four distinct cell types and two prominent Cell Painting datasets - RxRx1 and a subset of JUMP-CP. Across 484 classes of CRISPR perturbations in JUMP-CP our method improves accuracy by 15.7%.

cell biology↗

bioRxiv2025

Small molecule bioactivity benchmarks are often well-predicted by counting cells

Phenotypic profiling methods, such as Cell Painting and gene expression, have been widely used to predict compound bioactivity, often showing improvement over predictive models based on chemical structures alone. We discovered that a large subset of assays in widely-used benchmark datasets either directly relate to cell health and cytotoxicity or are assays intending to capture a more specific phenotype but whose active compounds impact cell count, while inactives do not. As a result, counting cells can achieve similar predictive performance as Cell Painting or gene expression data. Filtering benchmarks to include only assays relating to protein targets reveals that Cell Painting can capture information that cannot be predicted by mere cell counting. We re-evaluated three benchmark datasets used with Cell Painting data and observed that, in many cases, cell count models produced an AUC comparable to models using the full Cell Painting profiles. However, in protein-target-specific benchmarks across 17 distinct protein targets, Cell Painting features demonstrated unique predictive power, outperforming mean balanced accuracy from cell count models with a relative improvement of 19.6%. We propose five practical recommendations for benchmarking machine learning models for predicting bioactivity, including using cell count as a baseline feature. Although multi-class classification applications (such as matching samples based on their morphological profile) are less likely to be predictable by cell count than bioactivity benchmarks, these recommendations are broadly applicable to machine learning for drug discovery.

bioinformatics↗

bioRxiv2024

PertEval-scFM: Benchmarking Single-Cell Foundation Models for Perturbation Effect Prediction

In silico modeling of transcriptional responses to perturbations is crucial for advancing our understanding of cellular processes and disease mechanisms. We present PertEval-scFM, a standardized framework designed to evaluate models for perturbation effect prediction. We apply PertEval-scFM to benchmark zero-shot single-cell foundation model (scFM) embeddings against base-line models to assess whether these contextualized representations enhance perturbation effect prediction. Our results show that scFM embeddings do not provide consistent improvements over baseline models, especially under distribution shift. Overall, this study provides a systematic evaluation of zero-shot scFM embeddings for perturbation effect prediction, highlighting the challenges of this task and revealing the limitations of current-generation scFMs. Our findings underscore the need for specialized models and high-quality datasets that capture a broader range of cellular states. Source code and documentation can be found at: https://github.com/aaronwtr/PertEval.

bioinformatics↗

bioRxiv2023

Cell-Vision Fusion: A Swin Transformer-based Approach to Predicting Kinase Inhibitor Mechanism of Action from Cell Painting Data

Image-based profiling of the cellular response to drug compounds has proven to be an effective method to characterize the morphological changes resulting from chemical perturbation experiments. This approach has been useful in the field of drug discovery, ranging from phenotype-based screening to identifying a compounds mechanism of action or toxicity. As a greater amount of data becomes available however, there are growing demands for deep learning methods to be applied to perturbation data. In this paper we applied the transformer-based SwinV2 computer vision architecture to predict the mechanism of action of 10 kinase inhibitor compounds directly from raw images of the cellular response. This method outperforms the standard approach of using image-based profiles, multidimensional feature set representations generated by bioimaging software. Furthermore, we combined the best performing models for three different data modalities, raw images, image-based profiles and compound chemical structures, to form a fusion model, Cell-Vision Fusion (CVF). This approach classified the kinase inhibitors with 69.79% accuracy and 70.56% F1 score, 4.20% and 5.49% greater, respectively, than the best performing image-based profile method. Our work provides three techniques, specific to Cell Painting images, which enable the SwinV2 architecture to train effectively, and explores approaches to combat the significant batch effects present in large Cell Painting perturbation datasets.

cell biology↗

bioRxiv2023

Histopathological Domain Adaptation with Generative Adversarial Networks - Bridging the Domain Gap Between Thyroid Cancer Histopathology Datasets

Deep learning techniques are increasingly being used to classify medical imaging data with high accuracy. Despite this, due to often limited training data, these models can lack sufficient generalizability to predict unseen test data, produced in a different domain, with comparable performance. This study focuses on thyroid histopathology image classification and investigates whether a Generative Adversarial Network (GAN), trained with just 156 patient samples, can produce high quality synthetic images to sufficiently augment training data and improve overall model generalizability. Utilizing a StyleGAN2-ADA approach, the generative network produced images with an FID score of 5.05, matching state-of-the-art GAN results in non-medical domains with comparable dataset sizes. Augmenting the training data with these GAN-generated images increased model generalizability significantly when tested on external data, improving overall accuracy and F1 scores by 36% and 42% respectively. Most importantly, this performance improvement was observed on minority class images, tumour subtypes which are known to suffer from high levels of inter-observer variability when classified by trained pathologists.

bioinformatics↗

bioRxiv2021

LMPred: Predicting Antimicrobial Peptides Using Pre-Trained Language Models and Deep Learning

Antimicrobial peptides (AMPs) are increasingly being used in the development of new therapeutic drugs, in areas such as cancer therapy and hypertension. Additionally, they are seen as an alternative to antibiotics due to the increasing occurrence of bacterial resistance. Wet-laboratory experimental identification, however, is both time consuming and costly, so in-silico models are now commonly used in order to screen new AMP candidates. This paper proposes a novel approach of creating model inputs; using pre-trained language models to produce contextualized embeddings representing the amino acids within each peptide sequence, before a convolutional neural network is then trained as the classifier. The optimal model was validated on two datasets, being one previously used in AMP prediction research, and an independent dataset, created by this paper. Predictive accuracies of 93.33% and 88.26% were achieved respectively, outperforming all previous state-of-the-art classification models.

bioinformatics↗