Biology subjects

Cakiroglu, S. A.

Publications and source records attributed to Cakiroglu, S. A..

4 recordsLinked to original sources

bioRxiv2025

Multi-modal single-cell foundation models via dynamic token adaptation

AO_SCPLOWBSTRACTC_SCPLOWRecent advances in applying deep learning in genomics include DNA-language and single-cell foundation models. However, these models take only one data type as input. We introduce dynamic token adaptation and demonstrate how it combines these models to predict gene regulation at the single-cell level in different genetic contexts. Although the method is generalisable, we focus on an illustrative example by training an adapter from DNA-sequence embeddings to a single-cell foundation models token embedding space. As a qualitative evaluation, we assess the impact of DNA sequence changes on the models learned gene regulatory networks by mutating the transcriptional start site of the transcription factor GATA4 in silico, observing predicted expression changes in its target genes in fetal cardiomyocytes.

bioinformatics↗

bioRxiv2025

Detecting cell-level transcriptomic changes of Perturb-seq using Contrastive Fine-tuning of Single-Cell Foundation Models

Genome-scale perturbation cell atlases are an exciting new resource to understand the transcriptomic and phenotypic impact of single-gene activation or knockdown. However, in terms of differentially expressed genes identified, the signal detected in these data atlases is low, leading to the exclusion of most data from downstream analyses. Recent advances in single-cell foundation models have shown promise in capturing complex biological insights. However, their application to perturbation analysis, especially in predicting perturbed single-cell transcriptomes, remains limited. In this paper, we focus on learning representations of single-cell transcriptomes that capture subtle, yet important, transcriptome-wide changes, and we propose a novel fine-tuning strategy using contrastive learning to leverage single-cell foundation models for this task. We pre-train a single-cell foundation model and fine-tune on a genome-scale perturbation dataset using a contrastive loss, which minimises the distance between cell embeddings from unperturbed cells while maximising the distance between perturbed and unperturbed cells. We validate and test the model on unseen perturbations, demonstrating its ability to identify global biologically meaningful transcriptional changes not captured by traditional differential expression methods. Our approach provides a novel framework for analysing single-cell perturbation data and offers a more effective means of identifying perturbations that drive systemic gene expression changes.

bioinformatics↗

bioRxiv2021

ChromWave: Deciphering the DNA-encoded competition between transcription factors and nucleosomes with deep neural networks

Transcription factors (TFs) regulate gene expression by recognising and binding specific DNA sequences. At times, these regulatory elements may be occluded by nucleosomes, making them inaccessible for TF-binding. The competition for DNA occupancy between TFs and nucleosomes, and associated gene regulatory outputs, are important consequences of the cis-regulatory information encoded in the genome. However, these sequence patterns are subtle and remain difficult to interpret. Here, we introduce ChromWave, a deep-learning model that, for the first time, predicts the competing profiles for TF and nucleosomes occupancies with remarkable accuracy. Models trained using short- and long-fragment MNase-Seq data successfully learn the sequence preferences underlying TF and nucleosome occupancies across the entire yeast genome. They recapitulate nucleosome evictions from regions containing "strong" TF binding sites and knock-out simulations show nucleosomes gaining occupancy in the absence of these TFs, accompanied by lateral rearrangement of adjacent nucleosomes. At a local level, models anticipate with high accuracy the outcomes of detailed experimental analysis of partially unwrapped nucleosomes at the GAL4 UAS locus. Finally, we trained a ChromWave model that successfully predicts nucleosome positions at promoters in the human genome. We find that human promoters generally contain few sites at which simple sequence changes can alter nucleosome occupancies and that these positions align well with causal variants linked to DNase hypersensitivity.

genomics↗

bioRxiv2018

Machine learning models in electronic health records can outperform conventional survival models for predicting patient mortality in coronary artery disease

Prognostic modelling is important in clinical practice and epidemiology for patient management and research. Electronic health records (EHR) provide large quantities of data for such models, but conventional epidemiological approaches require significant researcher time to implement. Expert selection of variables, fine-tuning of variable transformations and interactions, and imputing missing values in datasets are time-consuming and could bias subsequent analysis, particularly given that missingness in EHR is both high, and may carry meaning.\n\nUsing a cohort of over 80,000 patients from the CALIBER programme, we performed a systematic comparison of several machine-learning approaches in EHR. We used Cox models and random survival forests with and without imputation on 27 expert-selected variables to predict all-cause mortality. We also used Cox models, random forests and elastic net regression on an extended dataset with 586 variables to build prognostic models and identify novel prognostic factors without prior expert input.\n\nWe observed that data-driven models used on an extended dataset can outperform conventional models for prognosis, without data preprocessing or imputing missing values, and with no need to scale or transform continuous data. An elastic net Cox regression based with 586 unimputed variables with continuous values discretised achieved a C-index of 0.801 (bootstrapped 95% CI 0.799 to 0.802), compared to 0.793 (0.791 to 0.794) for a traditional Cox model comprising 27 expert-selected variables with imputation for missing values.\n\nWe also found that data-driven models allow identification of novel prognostic variables; that the absence of values for particular variables carries meaning, and can have significant implications for prognosis; and that variables often have a nonlinear association with mortality, which discretised Cox models and random forests can elucidate.\n\nThis demonstrates that machine-learning approaches applied to raw EHR data can be used to build reliable models for use in research and clinical practice, and identify novel predictive variables and their effects to inform future research.

epidemiology↗