Zhaoyu Li

Background: Current methods for predicting gene expression from histone modifications rely on arbitrary binary classification thresholds such as the median to distinguish between high and low expression for genes. This approach lacks biological justification, creates dataset-dependent classifications, and ignores the relative regulatory relationships between genes that are often more biologically meaningful than absolute cutoffs. Methods: We developed a deep learning method using pairwise ranking to determine the relative gene expression levels between gene pairs based on their histone modification signals. The approach adopted the DirectRanker architecture, which enforces antisymmetry by construction through a twin-subnet design with asymmetric subtraction, and was evaluated against three baseline classifiers (Random Forest, Logistic Regression, and SVM Linear) on identical features and data splits. Ablation testing across all 31 non-empty subsets of five histone marks (H3K4me3, H3K9ac, H3K9me3, H3K27ac, H3K27me3) was conducted to identify their individual contributions. Datasets were strictly partitioned (80% training, 10% validation, 10% testing) with gene-level separation to prevent data leakage, and all experiments were repeated across five independent random seeds. Four datasets were analysed: two normal adult liver cell datasets (Donor 3 and Donor 4 from NCBI GEO GSE19465) and two HepG2 hepatocellular carcinoma cell line datasets (ENCSR134DWG and GSE76344). Results: Multi-mark combinations anchored by active histone marks (H3K27ac, H3K4me3, H3K9ac) consistently drove predictive performance, with the best combinations achieving AUROC values of 0.833−0.867 and test accuracies of 74–78% across datasets. DirectRanker attained AUPRC values of 0.819−0.856, closely matching or exceeding Random Forest on precision–recall performance. Repressive marks (H3K9me3, H3K27me3) consistently underperformed when used in isolation, with single-mark models achieving test accuracies of only 50–66%. DirectRanker produced substantially higher antisymmetry scores (0.961−0.983) than all baseline classifiers (0.787−0.851), confirming logically consistent pairwise predictions. Performance plateaued at four- to five-mark combinations, reflecting correlated predictive signal among the active marks within the promoter-proximal feature space. Conclusions: The pairwise ranking framework provides a principled alternative to threshold-based classification, enabling relative gene expression comparisons without arbitrary binarisation cutoffs. While Random Forest achieved marginally higher accuracy, DirectRanker’s structural antisymmetry guarantee makes it the more suitable choice for applications requiring a globally coherent gene ranking. Active promoter marks carry the majority of the predictive signal, while the performance saturation with increasing mark combinations reflects overlapping information content rather than biological redundancy.

Improving the efficiency of data compression remains essential for feature selection and data modelling. Current approaches for compressing epigenomic/genomic data highly rely on autoencoder that requires substantial computing resources, parameter fine-tuning, training, and time. Here, we developed a training-free, Fast Fourier Transform (FFT)-based method, for data compression with high efficiency and full interpretability. Our FFT method compresses epigenomic data of histone modification up to 1,000-fold while still maintaining high reconstruction fidelity (cosine similarity, 99.7%), does not require any training and completes ultrafast within 70 ms on GPU or 20 s on CPU opposite to extensive training in hours/days for autoencoder on GPU/CPU, and offers full interpretability of compressed features from frequency components of original signals in contrast to the uninterpretable “black box” from autoencoder. This enables high accuracy in the classification model prediction (AUC, 0.960). Thus, our novel FFT method represents a major paradigm shift in data compression.

Current methods for predicting gene expression from histone modifications rely on arbitrary binary classification thresholds such as the median to distinguish between high and low expression for genes. This approach lacks biological justification, creates dataset-dependent classifications, and ignores the relative regulatory relationships between genes that are often more biologically meaningful than absolute cutoffs. We introduce a novel pairwise ranking approach that compares relative expression levels between gene pairs based on their histone modification patterns, eliminating arbitrary threshold selection. Evaluation using the REMC E066 and GSE76344 datasets showed that our pairwise classification framework demonstrated consistent ranking performance in both datasets. The ablation study found that only a subset of histone modification types is necessary, giving evidence of redundancy in the biological code.

Improving the efficiency of data compression remains essential for feature selection and data modelling. Current approaches for compressing epigenomic/genomic data highly rely on autoencoder that requires substantial computing resources, parameter fine-tuning, training, and time. Here, we developed a training-free, Fast Fourier Transform (FFT)-based method, for data compression with high efficiency and full interpretability. Our FFT method compresses epigenomic data of histone modification up to 1,000-fold while still maintaining high reconstruction fidelity (cosine similarity, 99.7%), does not require any training and completes ultrafast within 70 milliseconds on GPU or 20 seconds on CPU opposite to extensive training in hours/days for autoencoder on GPU/CPU, and offers full interpretability of compressed features from frequency components of original signals in contrast to the uninterpretable “black box” from autoencoder. This enables high accuracy in the classification model prediction (AUC, 0.960). Thus, our novel FFT method represents a major paradigm shift in data compression.

Objectives: Studies on the impact of hormone replacement therapy (HRT) on hepatocellular carcinoma (HCC) have been investigated in the past. This study aims to further address this topic using a large population dataset of the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial (NIH, USA). Methods: Multivariable logistic regression was used for the analysis of women in the entire PLCO cohort. Results: A non-significant reduction in the risk of developing HCC was observed in women using HRT compared to never users (OR=0.61, 95 % CI: 0.21–1.79, p=0.367). Oral contraceptive use was significantly associated with reduced risk of HCC in the HRT group, (OR=0.17, 95 % CI: 0.00–0.82, p=0.028), and there was a nearly-significant interaction between HRT use and oral contraceptive use (p=0.054). Regardless of HRT use, those with pre-existing liver comorbidities of hepatitis or cirrhosis were at the highest risk of developing HCC (OR=5.09, 95 % CI: 1.13–22.81, p=0.034). However, liver comorbidities of hepatitis or cirrhosis and exposure to HRT showed a significant interaction (p=0.0001). Conclusions: In those taking HRT, oral contraceptive use was significantly protective against HCC, suggesting that a longer period of hormone use (oral contraceptive plus HRT) provides additional protection against HCC than HRT alone. Overall, the low numbers of HCC diagnoses, which reflect the wider healthy female population, limited our ability to establish statistical significance regarding the impact of HRT on HCC incidence, which requires further prospective studies of larger populations.

Summary: Glioblastoma and osteosarcoma originate from the same lineage, yet patients with these two tumour types show significant differences in survival outcomes. Transcriptomic analysis comparing these tumours reveals that over 65% genes show similar expression patterns. Principal component analysis further demonstrates substantial similarities between these two tumour types, albeit with discernible differences. Deep learning analysis employing an autoencoder unveils nuanced distinctions and similarities of these two tumours at a high resolution. A classification model, leveraging gradient boosting with eXtreme Gradient Boosting (XGBoost), achieves high accuracy in distinguishing between these two tumour types. Identification of key contributors to the model’s performance is facilitated by SHapley Additive exPlanations (SHAP), yielding two lists of top target genes with and without considering gender. Notably, these SHAP targets tend to cluster within one or two networks of signalling pathways. Remarkably, gene expression levels of many of these SHAP targets alone can recapitulate survival differences solely based on clinical data between glioblastoma and osteosarcoma patients. Of particular interest, C2ORF72 emerges as a common target from both lists, representing an uncharacterised protein with promising potential as a novel target for diagnostic, prognostic, and therapeutic target for both glioblastoma and osteosarcoma.

Breast cancer is a leading cause of cancer-related deaths among women. The identification of survival-related target genes is critical for improving the prognosis and outcomes of breast cancer patients. Many methods have been applied to this investigation, such as bioinformatics and machine learning approaches, yet few targets identified from these approaches have been applied in clinics. Here, we present a novel approach by using classical statistical methods of Kolmogorov-Smirnov (KS) test and Jensen-Shannon (JS) divergence to analyse the survival time and gene expression data of breast cancer patients (BRCA) from The Cancer Genome Atlas (TCGA). These methods help compare the survival time distributions and differentiate patients into high and low-risk groups based on gene expression profiles. 1,124 survival-related genes were identified based on the KS test and 18 from JS divergence values. We also identified the optimal thresholds of the expression level of these target genes, which enabled the best separation of survival groups for all breast cancer patients and each subtype of breast cancer patients. These targets were further validated through bootstrapping to ensure that significant results are not due to chance. By comparing those survival targets from previous studies, we found two were novel targets, and two were consistent with previous reports. Overall, our study provides a novel approach for identifying survival targets for breast cancer patients by integrating a series of classical statistical methods, such as the KS test, JS divergence, and bootstrapping. Our approach could also be applied to identifying the survival targets for other cancer types and provide valuable insights into cancer research and clinical applications.

Sex differences in the incidences of cancers become a critical issue in both cancer research and the development of precision medicine. However, details in these differences have not been well reported. We provide a comprehensive analysis of sexual dimorphism in human cancers. We analyzed four sets of cancer incidence data from the SEER (USA, 1975–2015), from the Cancer Registry at Mayo Clinic (1970–2015), from Sweden (1970–2015), and from the World Cancer Report in 2012. We found that all human cancers had statistically significant sexual dimorphism with male dominance in the United States and mostly significant in the Mayo Clinic, Sweden, and the world data, except for thyroid cancer, which is female-dominant. Sexual dimorphism is a clear but mostly neglected phenotype for most human cancers regarding the clinical practice of cancer. We expect that our study will facilitate the mechanistic studies of sexual dimorphism in human cancers. We believe that fully addressing the mechanisms of sexual dimorphism in human cancers will greatly benefit current development of individualized precision medicine beginning from the sex-specific diagnosis, prognosis, and treatment.