Jean Gao
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Jean Gao has a total of 80 co-authors. Here are the top co-authors by number of shared publications:
- Ashis Kumer Biswas
Computer Science & Engineering University of Colorado Denver 80234 Colorado CO
9 shared publications - Mingon Kang
Department of Computer Science, University of Nevada, Las Vegas, Las Vegas, USA
6 shared publications - Chunyu Liu
MOE Key Laboratory of Rare Pediatric Diseases & Hunan Key Laboratory of Medical Genetics, School of Life Sciences, and Department of Psychiatry, The Second Xiangya Hospital, Central South University, Changsha, China
5 shared publications - Dongchul Kim
Department of Computer Science, The University of Texas Rio Grande Valley, Edinburg, TX 78539, USA
5 shared publications - Liping Tang
Department of Bioengineering, University of Texas at Arlington, Arlington, Texas, USA
5 shared publications Xiaoyong Wu
School of Communications and Electronic Information, Tianjin Normal University
5 shared publications
Latest Publications
- Open Access
- 21 May 2026
- Published by Springer Nature in BMC Bioinformatics
- Vol. 27 (1)
Transcriptome graph transformer: a graph transformer-based unsupervised model for transcriptome data analysis
Rapidly growing transcriptomic datasets pose challenges for traditional analytical methods, which struggle with high dimensionality, heterogeneity, and nonlinear gene relationships. Existing deep learning models often require fixed-length inputs and fail to integrate biological network information. We introduce Transcriptome Graph Transformer (TGT), an unsupervised graph Transformer framework that constructs a heterogeneous gene-pathway graph using expression data, STRING interactions, and GO/KEGG/Reactome pathway annotations. TGT is pretrained with a masked-node prediction task and fine-tuned for disease classification, biomarker discovery, and zero-shot clustering of single-cell and spatial transcriptomics. TGT achieves superior performance across Alzheimer's disease, cancer, acute kidney injury, and COVID-19 datasets, outperforming state-of-the-art baselines. The model generalizes well across platforms and yields biologically meaningful gene and pathway importance scores consistent with known disease mechanisms. TGT provides an effective and generalizable approach for transcriptomic representation learning by integrating biological network knowledge with graph Transformer architectures. Its strong performance highlights its utility for broad transcriptomic applications and precision medicine.
- 21 May 2026
- Published by Springer Nature in BMC Bioinformatics
- Vol. 27 (1)
Open AccessTranscriptome graph transformer: a graph transformer-based unsupervised model for transcriptome data analysis
Rapidly growing transcriptomic datasets pose challenges for traditional analytical methods, which struggle with high dimensionality, heterogeneity, and nonlinear gene relationships. Existing deep learning models often require fixed-length inputs and fail to integrate biological network information. We introduce Transcriptome Graph Transformer (TGT), an unsupervised graph Transformer framework that constructs a heterogeneous gene-pathway graph using expression data, STRING interactions, and GO/KEGG/Reactome pathway annotations. TGT is pretrained with a masked-node prediction task and fine-tuned for disease classification, biomarker discovery, and zero-shot clustering of single-cell and spatial transcriptomics. TGT achieves superior performance across Alzheimer's disease, cancer, acute kidney injury, and COVID-19 datasets, outperforming state-of-the-art baselines. The model generalizes well across platforms and yields biologically meaningful gene and pathway importance scores consistent with known disease mechanisms. TGT provides an effective and generalizable approach for transcriptomic representation learning by integrating biological network knowledge with graph Transformer architectures. Its strong performance highlights its utility for broad transcriptomic applications and precision medicine.
- Open Access
- 22 January 2026
- Published by Springer Nature in Research Square
DAG-VAERL: a novel causal inference method for building causal gene regulatory network
Causal discovery methods provide a powerful tool for uncovering the causal relationships between lncRNAs and target genes in gene regulations. In many causal inference and structure learning tasks, learning the Directed Acyclic Graph (DAG) structure from data is a challenging problem. Traditional DAG learning methods often rely on heuristic searches or strict constraints, which fail to effectively handle complex nonlinear relationships and discrete data.To address this, we propose a novel deep generative model — DAG-VAERL, which combines Graph Neural Networks (GNN) as well as Reinforcement Learning (RL) frameworks and Graph Attention Networks (GAT) module, leveraging Variational Autoencoders (VAE) to learn the DAG structure. DAG-VAERL is capable of modeling complex dependencies between nodes through GNNs and optimizing the graph structure using RL strategies. We conduct extensive experiments on synthetic and real-world datasets, including Alzheimer's disease data, to validate the superiority of DAG-VAERL in structural discovery and parameter estimation. Experimental results demonstrate that DAG-VAERL significantly outperforms traditional methods in structure recovery, especially when dealing with complex data involving nonlinear and discrete variables. This model not only effectively learns the DAG structure from data but also serves as a powerful tool for causal inference and other graph-bard analysis tasks, providing a new approach for related fields.
- 22 January 2026
- Published by Springer Nature in Research Square
Open AccessDAG-VAERL: a novel causal inference method for building causal gene regulatory network
Causal discovery methods provide a powerful tool for uncovering the causal relationships between lncRNAs and target genes in gene regulations. In many causal inference and structure learning tasks, learning the Directed Acyclic Graph (DAG) structure from data is a challenging problem. Traditional DAG learning methods often rely on heuristic searches or strict constraints, which fail to effectively handle complex nonlinear relationships and discrete data.To address this, we propose a novel deep generative model — DAG-VAERL, which combines Graph Neural Networks (GNN) as well as Reinforcement Learning (RL) frameworks and Graph Attention Networks (GAT) module, leveraging Variational Autoencoders (VAE) to learn the DAG structure. DAG-VAERL is capable of modeling complex dependencies between nodes through GNNs and optimizing the graph structure using RL strategies. We conduct extensive experiments on synthetic and real-world datasets, including Alzheimer's disease data, to validate the superiority of DAG-VAERL in structural discovery and parameter estimation. Experimental results demonstrate that DAG-VAERL significantly outperforms traditional methods in structure recovery, especially when dealing with complex data involving nonlinear and discrete variables. This model not only effectively learns the DAG structure from data but also serves as a powerful tool for causal inference and other graph-bard analysis tasks, providing a new approach for related fields.
- Open Access
- 9 January 2026
- Published by Springer Nature in Research Square
Transcriptome Graph Transformer--A Graph Transformer-Based Unsupervised Model for Transcriptome Data Analysis
Background: Rapidly growing transcriptomic datasets pose challenges for traditional analytical methods, which struggle with high dimensionality, heterogeneity, and nonlinear gene relationships. Existing deep learning models often require fixed-length inputs and fail to integrate biological network information. Methods: We introduce Transcriptome Graph Transformer (TGT), an unsupervised graph Transformer framework that constructs a heterogeneous gene--pathway graph using expression data, STRING interactions, and GO/KEGG/Reactome pathway annotations. TGT is pretrained with a masked-node prediction task and fine-tuned for disease classification, biomarker discovery, and zero-shot clustering of single-cell and spatial transcriptomics. Results: TGT achieves superior performance across Alzheimer's disease, cancer, acute kidney injury, and COVID-19 datasets, outperforming state-of-the-art baselines. The model generalizes well across platforms and yields biologically meaningful gene and pathway importance scores consistent with known disease mechanisms. Conclusion: TGT provides an effective and generalizable approach for transcriptomic representation learning by integrating biological network knowledge with graph Transformer architectures. Its strong performance highlights its utility for broad transcriptomic applications and precision medicine.
- 9 January 2026
- Published by Springer Nature in Research Square
Open AccessTranscriptome Graph Transformer--A Graph Transformer-Based Unsupervised Model for Transcriptome Data Analysis
Background: Rapidly growing transcriptomic datasets pose challenges for traditional analytical methods, which struggle with high dimensionality, heterogeneity, and nonlinear gene relationships. Existing deep learning models often require fixed-length inputs and fail to integrate biological network information. Methods: We introduce Transcriptome Graph Transformer (TGT), an unsupervised graph Transformer framework that constructs a heterogeneous gene--pathway graph using expression data, STRING interactions, and GO/KEGG/Reactome pathway annotations. TGT is pretrained with a masked-node prediction task and fine-tuned for disease classification, biomarker discovery, and zero-shot clustering of single-cell and spatial transcriptomics. Results: TGT achieves superior performance across Alzheimer's disease, cancer, acute kidney injury, and COVID-19 datasets, outperforming state-of-the-art baselines. The model generalizes well across platforms and yields biologically meaningful gene and pathway importance scores consistent with known disease mechanisms. Conclusion: TGT provides an effective and generalizable approach for transcriptomic representation learning by integrating biological network knowledge with graph Transformer architectures. Its strong performance highlights its utility for broad transcriptomic applications and precision medicine.
- Open Access
- 11 December 2025
- Published by Springer Nature in Research Square
Attention-Based Hierarchical Graph Autoencoder for Dose-Specific Single-Cell Resistance Dynamics
Background: Understanding cellular adaptation to drug concentrations is critical for modeling therapeutic resistance. Current perturbation models treat doses as categorical variables and lack biological hierarchy, limiting their ability to model continuous dose-response trajectories and capture pathway-level regulatory rewiring. Results: We introduce a hierarchical graph attention autoencoder that models dose-specific single-cell transcriptional responses by integrating pathway, gene, and cell hierarchies with Long Short-Term Memory (LSTM)-based dose conditioning. Applied to GSE206125 (BRCA2-deficient ovarian cancer cells exposed to escalating olaparib concentrations T1–T320), our model reconstructs multiscale embeddings while predicting resistance phenotypes. Five-fold cross-validation with held-out doses demonstrates robust dose-level generalization (validation loss: 0.160 ± 0.045). The model extrapolates to unseen dose T320 with high fidelity (cosine similarity: 0.98) and successfully predicts trajectory responses at T640, beyond the training range. LSTM-based encoding captures non-linear transcriptional dynamics, including regime transitions between T40 and T80. Multi-head attention mechanisms consistently prioritize resistance-associated pathways (TP53-regulated transcription, oxidative stress signaling) and key regulatory genes (CDKN2A, YBX1). Ablation studies confirm that virtual nodes, attention mechanisms, and LSTM dose encoding each contribute uniquely to reconstruction fidelity and biological coherence. Conclusions: Our framework enables interpretable modeling of dose-dependent transcriptional trajectories with demonstrated extrapolation capability. While applied here to PARP inhibitor response in ovarian cancer, the architecture is designed to accommodate other dose-escalation studies as appropriately structured datasets become available.
- 11 December 2025
- Published by Springer Nature in Research Square
Open AccessAttention-Based Hierarchical Graph Autoencoder for Dose-Specific Single-Cell Resistance Dynamics
Background: Understanding cellular adaptation to drug concentrations is critical for modeling therapeutic resistance. Current perturbation models treat doses as categorical variables and lack biological hierarchy, limiting their ability to model continuous dose-response trajectories and capture pathway-level regulatory rewiring. Results: We introduce a hierarchical graph attention autoencoder that models dose-specific single-cell transcriptional responses by integrating pathway, gene, and cell hierarchies with Long Short-Term Memory (LSTM)-based dose conditioning. Applied to GSE206125 (BRCA2-deficient ovarian cancer cells exposed to escalating olaparib concentrations T1–T320), our model reconstructs multiscale embeddings while predicting resistance phenotypes. Five-fold cross-validation with held-out doses demonstrates robust dose-level generalization (validation loss: 0.160 ± 0.045). The model extrapolates to unseen dose T320 with high fidelity (cosine similarity: 0.98) and successfully predicts trajectory responses at T640, beyond the training range. LSTM-based encoding captures non-linear transcriptional dynamics, including regime transitions between T40 and T80. Multi-head attention mechanisms consistently prioritize resistance-associated pathways (TP53-regulated transcription, oxidative stress signaling) and key regulatory genes (CDKN2A, YBX1). Ablation studies confirm that virtual nodes, attention mechanisms, and LSTM dose encoding each contribute uniquely to reconstruction fidelity and biological coherence. Conclusions: Our framework enables interpretable modeling of dose-dependent transcriptional trajectories with demonstrated extrapolation capability. While applied here to PARP inhibitor response in ovarian cancer, the architecture is designed to accommodate other dose-escalation studies as appropriately structured datasets become available.
- Top Cited Papers
- 27 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 36 (12) , 19986-19995
Segment Any Cell: A SAM-Based Auto-Prompting Fine-Tuning Framework for Nuclei Segmentation
In the rapidly evolving field of AI research, foundational models like BERT and GPT have significantly advanced language and vision tasks. The advent of pretrain-prompting models, such as ChatGPT and segment anything model (SAM), has further revolutionized image segmentation. However, their applications in specialized areas, particularly in nuclei segmentation within medical imaging, reveal a key challenge: the generation of high-quality, informative prompts is as crucial as applying state-of-the-art (SOTA) fine-tuning techniques on foundation models. To address this, we introduce segment any cell (SAC), an innovative framework that enhances SAM specifically for nuclei segmentation. SAC integrates a low-rank adaptation (LoRA) within the attention layer of the Transformer to improve the fine-tuning process, outperforming existing SOTA methods. It also introduces an innovative auto-prompt generator that produces effective prompts to guide segmentation, a critical factor in handling the complexities of nuclei segmentation in biomedical imaging. Our extensive experiments demonstrate the superiority of SAC in nuclei segmentation tasks, proving its effectiveness as a tool for pathologists and researchers. Our contributions include a novel prompt generation strategy, automated adaptability for diverse segmentation tasks, the innovative application of low-rank attention adaptation in SAM, and a versatile framework for semantic and instance segmentation challenges.
- 27 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 36 (12) , 19986-19995
Top Cited PapersSegment Any Cell: A SAM-Based Auto-Prompting Fine-Tuning Framework for Nuclei Segmentation
In the rapidly evolving field of AI research, foundational models like BERT and GPT have significantly advanced language and vision tasks. The advent of pretrain-prompting models, such as ChatGPT and segment anything model (SAM), has further revolutionized image segmentation. However, their applications in specialized areas, particularly in nuclei segmentation within medical imaging, reveal a key challenge: the generation of high-quality, informative prompts is as crucial as applying state-of-the-art (SOTA) fine-tuning techniques on foundation models. To address this, we introduce segment any cell (SAC), an innovative framework that enhances SAM specifically for nuclei segmentation. SAC integrates a low-rank adaptation (LoRA) within the attention layer of the Transformer to improve the fine-tuning process, outperforming existing SOTA methods. It also introduces an innovative auto-prompt generator that produces effective prompts to guide segmentation, a critical factor in handling the complexities of nuclei segmentation in biomedical imaging. Our extensive experiments demonstrate the superiority of SAC in nuclei segmentation tasks, proving its effectiveness as a tool for pathologists and researchers. Our contributions include a novel prompt generation strategy, automated adaptability for diverse segmentation tasks, the innovative application of low-rank attention adaptation in SAM, and a versatile framework for semantic and instance segmentation challenges.
- Open Access
- 22 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-8
MFMF: Multiple Foundation Model Fusion Networks for Whole Slide Image Classification
Tumor detection and subtyping remain a significant challenge in histopathology image analysis. As digital pathology progresses, the applications of deep learning become essential. Whole Slide Image (WSI) classification has emerged as a crucial task in digital pathology, vital for accurate cancer diagnosis and treatment. In this paper, we introduce an innovative abnormal-guided Multiple Foundation Model Fusion (MFMF) framework, aimed at enhancing WSI classification by integrating multi-level information from pathology images with Multiple Instance Learning (MIL). Traditional methods often focus on patch-level features while neglecting the rich contextual and morphological details at the cell and text levels, thus failing to fully exploit the multidimensional nature of WSIs. Our method enhances traditional models by efficiently integrating patch-level, cell-level, and text-level features using three foundation models. These are then fused through a novel three-step cross-attention module that effectively leverages cell and text information with patch-level features. Furthermore, unlike most studies that use attention scores to select instances based on the assumption that high scores indicate the presence of a tumor, we design an abnormality-aware module to naturally identify and detect abnormal features (i.e., tumors) as the criteria for selecting important instances, thereby reducing computational costs and boosting overall performance. We validate our approach against leading benchmarks on the CAMELYON16 and TCGA-Lung datasets, achieving superior classification performance. Our study not only tackles the challenges of sparsity and noise in multi-level features but also enhances the efficiency and accuracy of WSI classification by exploiting abnormal features.
- 22 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-8
Open AccessMFMF: Multiple Foundation Model Fusion Networks for Whole Slide Image Classification
Tumor detection and subtyping remain a significant challenge in histopathology image analysis. As digital pathology progresses, the applications of deep learning become essential. Whole Slide Image (WSI) classification has emerged as a crucial task in digital pathology, vital for accurate cancer diagnosis and treatment. In this paper, we introduce an innovative abnormal-guided Multiple Foundation Model Fusion (MFMF) framework, aimed at enhancing WSI classification by integrating multi-level information from pathology images with Multiple Instance Learning (MIL). Traditional methods often focus on patch-level features while neglecting the rich contextual and morphological details at the cell and text levels, thus failing to fully exploit the multidimensional nature of WSIs. Our method enhances traditional models by efficiently integrating patch-level, cell-level, and text-level features using three foundation models. These are then fused through a novel three-step cross-attention module that effectively leverages cell and text information with patch-level features. Furthermore, unlike most studies that use attention scores to select instances based on the assumption that high scores indicate the presence of a tumor, we design an abnormality-aware module to naturally identify and detect abnormal features (i.e., tumors) as the criteria for selecting important instances, thereby reducing computational costs and boosting overall performance. We validate our approach against leading benchmarks on the CAMELYON16 and TCGA-Lung datasets, achieving superior classification performance. Our study not only tackles the challenges of sparsity and noise in multi-level features but also enhances the efficiency and accuracy of WSI classification by exploiting abnormal features.
- Open Access
- 22 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-8
AlphaEpi: Enhancing B Cell Epitope Prediction with AlphaFold 3
Accurately identifying conformational B-cell epitopes, which involve complex protein structures, is essential in modern immunology and vaccine development. Traditional methods like X-ray crystallography provide precise data but are costly and time-consuming. This has led to the rise of computational methods, especially deep learning algorithms, which offer more efficient and cost-effective solutions. A major challenge in predicting conformational epitopes based on structure is the limited availability of experimentally validated structural data. Structural prediction tools can help address this issue but still rely on provided sequence information and complex algorithms, which can introduce noise and affect accuracy. Moreover, effectively aligning and fusing structural features with evolutionary sequence information remains another challenge. To address these challenges, we introduce AlphaEpi, an innovative model that combines the advanced structure prediction capabilities of AlphaFold 3 with deep learning graph neural networks. Additionally, we propose for the first time a dynamic selector module that allows the model to adjust its processing strategy based on the reliability of the structure and sequence information, addressing the challenge posed by unreliable structural data. Moreover, a novel graph fusion module integrates structural and evolutionary information, effectively solving the disparities in the integration process of structure and sequence data. Results show that by integrating advanced structural prediction technologies AlphaFold 3 with deep learning graph neural networks, AlphaEpi significantly surpasses baseline results, establishing itself as the state-of-the-art. This advancement furthers the development of B-cell epitope prediction and has significant implications for biological research.
- 22 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-8
Open AccessAlphaEpi: Enhancing B Cell Epitope Prediction with AlphaFold 3
Accurately identifying conformational B-cell epitopes, which involve complex protein structures, is essential in modern immunology and vaccine development. Traditional methods like X-ray crystallography provide precise data but are costly and time-consuming. This has led to the rise of computational methods, especially deep learning algorithms, which offer more efficient and cost-effective solutions. A major challenge in predicting conformational epitopes based on structure is the limited availability of experimentally validated structural data. Structural prediction tools can help address this issue but still rely on provided sequence information and complex algorithms, which can introduce noise and affect accuracy. Moreover, effectively aligning and fusing structural features with evolutionary sequence information remains another challenge. To address these challenges, we introduce AlphaEpi, an innovative model that combines the advanced structure prediction capabilities of AlphaFold 3 with deep learning graph neural networks. Additionally, we propose for the first time a dynamic selector module that allows the model to adjust its processing strategy based on the reliability of the structure and sequence information, addressing the challenge posed by unreliable structural data. Moreover, a novel graph fusion module integrates structural and evolutionary information, effectively solving the disparities in the integration process of structure and sequence data. Results show that by integrating advanced structural prediction technologies AlphaFold 3 with deep learning graph neural networks, AlphaEpi significantly surpasses baseline results, establishing itself as the state-of-the-art. This advancement furthers the development of B-cell epitope prediction and has significant implications for biological research.
- Open Access
Solving All Regression Models For Learning Gaussian Networks Using Givens Rotations
Score based learning (SBL) is a promising approach for learning Bayesian networks. The initial step in the majority of the SBL algorithms consists of computing the scores of all possible child and parent-set combinations for the variables. For Bayesian networks with continuous variables, a particular score is usually calculated as a function of the regression of the child over the variables in the parent-set. The sheer number of regressions models to be solved necessitates the design of efficient numerical algorithms. In this paper, we propose an algorithm for an efficient and exact calculation of regressions for all child and parent-set combinations. In the proposed algorithm, we use QR decompositions (QRDs) to capture the dependencies between the regressions for different families and Givens rotations to efficiently traverse through the space of QRDs such that all the regression models are accounted for in the shortest path possible. We compare the complexity of the suggested method with different algorithms, mainly those arising in all subset regression problems, and show that our algorithm has the smallest algorithmic complexity. We also explain how to parallelize the proposed method so as to decrease the runtime by a factor proportional to the number of processors utilized.
Open AccessSolving All Regression Models For Learning Gaussian Networks Using Givens Rotations
Score based learning (SBL) is a promising approach for learning Bayesian networks. The initial step in the majority of the SBL algorithms consists of computing the scores of all possible child and parent-set combinations for the variables. For Bayesian networks with continuous variables, a particular score is usually calculated as a function of the regression of the child over the variables in the parent-set. The sheer number of regressions models to be solved necessitates the design of efficient numerical algorithms. In this paper, we propose an algorithm for an efficient and exact calculation of regressions for all child and parent-set combinations. In the proposed algorithm, we use QR decompositions (QRDs) to capture the dependencies between the regressions for different families and Givens rotations to efficiently traverse through the space of QRDs such that all the regression models are accounted for in the shortest path possible. We compare the complexity of the suggested method with different algorithms, mainly those arising in all subset regression problems, and show that our algorithm has the smallest algorithmic complexity. We also explain how to parallelize the proposed method so as to decrease the runtime by a factor proportional to the number of processors utilized.
- Open Access
Network Representation of Large-Scale Heterogeneous RNA Sequences with Integration of Diverse Multi-omics, Interactions, and Annotations Data
Long non-coding RNA, microRNA, and messenger RNA enable key regulations of various biological processes through a variety of diverse interaction mechanisms. Identifying the interactions and cross-talk between these heterogeneous RNA classes is essential in order to uncover the functional role of individual RNA transcripts, especially for unannotated and newly-discovered RNA sequences with no known interactions. Recently, sequence-based deep learning and network embedding methods are becoming promising approaches that can either predict RNA-RNA interactions from a sequence or infer missing interactions from patterns that may exist in the network topology. However, the majority of these methods have several limitations, eg, the inability to perform inductive predictions, to distinguish the directionality of interactions, or to integrate various sequence, interaction, and annotation biological datasets. We proposed a novel deep learning-based framework, rna2rna, which learns from RNA sequences to produce a low-dimensional embedding that preserves the proximities in both the interactions topology and the functional affinity topology. In this proposed embedding space, we have designated a two-part" source and target contexts" to capture the targeting and receptive fields of each RNA transcript, while encapsulating the heterogenous cross-talk interactions between lncRNAs and miRNAs. From experimental results, our method exhibits superior performance in AUPR rates compared to state-of-art approaches at predicting missing interactions in different RNA-RNA interaction databases and was shown to accurately perform link predictions to novel RNA sequences not seen at training time, even without any prior information. Additional results suggest that our proposed framework can capture a manifold for heterogeneous RNA sequences to discover novel functional annotations.
Open AccessNetwork Representation of Large-Scale Heterogeneous RNA Sequences with Integration of Diverse Multi-omics, Interactions, and Annotations Data
Long non-coding RNA, microRNA, and messenger RNA enable key regulations of various biological processes through a variety of diverse interaction mechanisms. Identifying the interactions and cross-talk between these heterogeneous RNA classes is essential in order to uncover the functional role of individual RNA transcripts, especially for unannotated and newly-discovered RNA sequences with no known interactions. Recently, sequence-based deep learning and network embedding methods are becoming promising approaches that can either predict RNA-RNA interactions from a sequence or infer missing interactions from patterns that may exist in the network topology. However, the majority of these methods have several limitations, eg, the inability to perform inductive predictions, to distinguish the directionality of interactions, or to integrate various sequence, interaction, and annotation biological datasets. We proposed a novel deep learning-based framework, rna2rna, which learns from RNA sequences to produce a low-dimensional embedding that preserves the proximities in both the interactions topology and the functional affinity topology. In this proposed embedding space, we have designated a two-part" source and target contexts" to capture the targeting and receptive fields of each RNA transcript, while encapsulating the heterogenous cross-talk interactions between lncRNAs and miRNAs. From experimental results, our method exhibits superior performance in AUPR rates compared to state-of-art approaches at predicting missing interactions in different RNA-RNA interaction databases and was shown to accurately perform link predictions to novel RNA sequences not seen at training time, even without any prior information. Additional results suggest that our proposed framework can capture a manifold for heterogeneous RNA sequences to discover novel functional annotations.