Dinghao Wang

Dinghao Wang

Auto-generated profile, based on publicly available information

Dinghao Wang

University of Calgary

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
15 September 2023 in bioRxiv
Verified
Preprint
Open access
PDF file
QLQing Li
Qing Li
DPDeshan Perera
Deshan Perera
ZCZhishan Chen
Zhishan Chen
WWWanqing Wen
Wanqing Wen
DWDinghao Wang
Dinghao Wang
QLQuan Long
Quan Long
Machine learning (ML) has proven successful in biological data analysis. However, may require massive training data. To allow broader use of ML in the full spectrum of biology and medicine, including sample-sparse domains, re-directing established models to specific tasks by add-on training via a moderate sample may be promising. Transfer learning (TL), a technique migrating pre-trained models to new tasks, fits in this requirement. Here, by TL, we retasked Enformer, a comprehensive model trained by massive data, tailored to breast cancers using breast-specific data. Its performance has been validated through statistical accuracy of predictions, annotation of genetic variants, and mapping of variants associated with breast cancer. By allowing the flexibility of adding dedicated training data, our TL protocol unlocks future discovery within specific domains with moderate add-on samples by standing on the shoulders of giant models.

5 October 2023 in bioRxiv
Verified
Preprint
Open access
PDF file
YQYanzhao Qian
Yanzhao Qian
DWDinghao Wang
Dinghao Wang
QDQi Xuan Ding
Qi Xuan Ding
MGMatthew Greenberg
Matthew Greenberg
QLQuan Long
Quan Long
Cross-validation (CV) is a widely used technique in statistical learning for model evaluation and selection. Meanwhile, various of statistical learning methods, such as Generalized Least Square (GLS), Linear Mixed-Effects Models (LMM), and regularization methods are commonly used in genomic predictions, a field that utilizes DNA polymorphisms to predict phenotypic traits. However, due to high dimensionality, relatively small sample sizes, and data sparsity in genomic data, CV in these scenarios may lead to an underestimation of the generalization error. In this work, we analyzed the bias of CV in eight methods: Ordinary Least Square (OLS), GLS, LMM, Lasso, Ridge, elastic-net (ENET), and two hybrid methods: one combining GLS with Ridge regularization (GLS+Ridge), and the other combining LMM with Ridge regularization (LMM+Ridge). Leveraging genomics data from the 1,000 Genomes Project and simulated phenotypes, our investigation revealed the presence of bias in all these methods. To address this bias, we adapted a variance-structure method known as Cross-Validation Correction (CVc). This approach aims to rectify the cross-validation error by providing a more accurate estimate of the generalization error. To quantify the performance of our adapted CVc towards all these methods, we applied the trained model to an independently generated dataset, which served as a gold standard for validating the models and calculating the generalization error. The outcomes show that, by leveraging CVc, we corrected the CV bias for most of the methods mentioned above, with two exceptions that are unrectifiable methods: ENET and Lasso. Our work revealed the substantial bias in the use of CV in genomics, a phenomenon under-appreciated by the field of statistical genomics and medicine. Additionally, we demonstrated that bias-corrected models may be formed by adapting CVc, although more work is needed to cover the full spectrum.

29 July 2023 in bioRxiv
Verified
Preprint
Open access
PDF file
HHHamid Hamidi
Hamid Hamidi
DWDinghao Wang
Dinghao Wang
QLQuan Long
Quan Long
MGMatthew Greenberg
Matthew Greenberg
Deep learning has been widely considered more effective than traditional statistical models in modeling biological complex data such as single-cell omics. Here we show the devil is hidden in details: by adapting a modern gradient solver to a traditional linear mixed model, we showed that conventional models can outperform deep models in terms of both speed and accuracy. This work reveals the potential of re-implementing traditional models with modern solvers.

6 May 2026 in PLOS Genetics
Verified
Journal Article
Open access
PDF file
QLQing Li
Qing Li
DWDinghao Wang
Dinghao Wang
ZZZilong Zhang
Zilong Zhang
DPDeshan Perera
Deshan Perera
ZCZhishan Chen
Zhishan Chen
QLQuan Long
Quan Long
DNA foundation models trained on large-scale genomic and epigenetic datasets have shown promise for regulatory variant interpretation, yet their application to tissue-specific contexts remain limited. Here, we present a transfer learning (TL) framework to adapt Enformer, a deep neural network trained on 5,313 multi-omics tracks, to breast and prostate cancer using 275 and 357 tissue-specific transcription factor (TF) ChIP–seq tracks, respectively. We computed tissue-specific cis-regulatory activity (tCRA) scores for millions of single-nucleotide variants (SNVs) in genome-wide association study (GWAS) datasets and prioritized high-impact SNV subsets (1M, 1.5M, and 2M). These TL-prioritized variants demonstrated consistently greater enrichment in tissue-specific enhancers, cancer GWAS risk variants, and ClinVar pathogenic variants compared to the original Enformer model. Transcriptome-wide association studies (TWAS) using TL-based SNVs identified more cancer-relevant genes, many of which exhibited functional essentiality (DepMap), therapeutic tractability (drug databases), and disease relevance (DisGeNET). Notably, TL models outperformed the base model in identifying genes enriched for drug targets and clinically relevant disease associations. Our results show that TL-derived tCRA scores enhance regulatory variant prioritization and improve susceptibility gene discovery in a tissue-specific manner. Our study provides a generalizable framework for tailoring foundation models to disease-relevant contexts, with implications for variant interpretation, therapeutic target discovery, and precision medicine.

17 August 2026 in PLOS Computational Biology
Verified
Journal Article
Open access
DWDinghao Wang
Dinghao Wang
PKPathum Kossinna
Pathum Kossinna
KAKaren Ardila
Karen Ardila
SKSenitha Kumarapeli
Senitha Kumarapeli
MMM. Ethan MacDonald
M. Ethan MacDonald
QZQingrun Zhang
Qingrun Zhang
Genetic contributions to complex traits are often mediated through coordinated gene–gene interaction networks, yet most existing association frameworks focus on marginal single-gene effects and overlook higher-order dependency structures. Direct modeling of interactions remains challenging due to combinatorial complexity and statistical instability. We introduce Interaction-Bridged Association Study (IBAS), a general framework that incorporates pathway-level interaction patterns into genotype–phenotype association analysis without explicitly enumerating interactions. IBAS leverages transcriptomic reference data to construct low-dimensional representations of pathway activity, which guide SNP-weighting and gene-level association testing within a kernel-based framework. In perturbation-based simulations, IBAS demonstrates improved stability and reproducibility compared to conventional TWAS and gene-based methods, while maintaining well-calibrated Type I error under phenotype permutation. Application to the WTCCC datasets identifies both known and novel genes across multiple complex diseases, including candidates with modest marginal effects missed by standard approaches. These findings are supported by replication in an independent cohort, and analyses across multiple reference tissues revealing both shared and tissue-specific signals. Overall, IBAS provides a statistically robust and computationally tractable framework for incorporating interaction effects into association mapping, extending beyond the single-gene paradigm and enabling more comprehensive characterization of complex trait. IBAS is available on GitHub at: https://github.com/QingrunZhangLab/IBAS

19 February 2022 in bioRxiv
Verified
Preprint
Open access
PDF file
DWDinghao Wang
Dinghao Wang
JHJingni He
Jingni He
DPDeshan Perera
Deshan Perera
CCChen Cao
Chen Cao
PKPathum Kossinna
Pathum Kossinna
QZQingrun Zhang
Qingrun Zhang
Linkage disequilibrium (LD) is a fundamental concept in genetics; critical for studying genetic associations and molecular evolution. However, LD measurements are only reliable for common genetic variants, leaving low-frequency variants unanalyzed. In this work, we introduce cumulative LD (cLD), a stable statistic that captures the rare-variant LD between genetic regions, which reflects more biological interactions between variants, in addition to lack of recombination. We derived the theoretical variance of cLD using delta methods to demonstrate its higher stability than LD for rare variants. This property is also verified by bootstrapped simulations using real data. In application, we find cLD reveals an increased genetic association between genes in 3D chromatin interactions, a phenomenon recently reported negatively by calculating standard LD between common variants. Additionally, we show that cLD is higher between gene pairs reported in interaction databases, identifies unreported protein-protein interactions, and reveals interacting genes distinguishing case/control samples in association studies.