HELM-GPT: de novo macrocyclic peptide design using generative pre-trained transformer.

MOTIVATION: Macrocyclic peptides hold great promise as therapeutics targeting intracellular proteins. This stems from their remarkable ability to bind flat protein surfaces with high affinity and specificity while potentially traversing the cell membrane. Research has already explored their use in developing inhibitors for intracellular proteins, such as KRAS, a well-known driver in various cancers. However, computational approaches for de novo macrocyclic peptide design remain largely unexplored. RESULTS: Here, we introduce HELM-GPT, a novel method that combines the strength of the hierarchical editing language for macromolecules (HELM) representation and generative pre-trained transformer (GPT) for de novo macrocyclic peptide design. Through reinforcement learning (RL), our experiments demonstrate that HELM-GPT has the ability to generate valid macrocyclic peptides and optimize their properties. Furthermore, we introduce a contrastive preference loss during the RL process, further enhanced the optimization performance. Finally, to co-optimize peptide permeability and KRAS binding affinity, we propose a step-by-step optimization strategy, demonstrating its effectiveness in generating molecules fulfilling both criteria. In conclusion, the HELM-GPT method can be used to identify novel macrocyclic peptides to target intracellular proteins. AVAILABILITY AND IMPLEMENTATION: The code and data of HELM-GPT are freely available on GitHub (https://github.com/charlesxu90/helm-gpt).

GAGE-seq concurrently profiles multiscale 3D genome organization and gene expression in single cells.

The organization of mammalian genomes features a complex, multiscale three-dimensional (3D) architecture, whose functional significance remains elusive because of limited single-cell technologies that can concurrently profile genome organization and transcriptional activities. Here, we introduce genome architecture and gene expression by sequencing (GAGE-seq), a scalable, robust single-cell co-assay measuring 3D genome structure and transcriptome simultaneously within the same cell. Applied to mouse brain cortex and human bone marrow CD34+ cells, GAGE-seq characterized the intricate relationships between 3D genome and gene expression, showing that multiscale 3D genome features inform cell-type-specific gene expression and link regulatory elements to target genes. Integration with spatial transcriptomic data revealed in situ 3D genome variations in mouse cortex. Observations in human hematopoiesis unveiled discordant changes between 3D genome organization and gene expression, underscoring a complex, temporal interplay at the single-cell level. GAGE-seq provides a powerful, cost-effective approach for exploring genome structure and gene expression relationships at the single-cell level across diverse biological contexts.

scGHOST: identifying single-cell 3D genome subcompartments.

Single-cell Hi-C (scHi-C) technologies allow for probing of genome-wide cell-to-cell variability in three-dimensional (3D) genome organization from individual cells. Computational methods have been developed to reveal single-cell 3D genome features based on scHi-C, including A/B compartments, topologically associating domains and chromatin loops. However, no method exists for annotating single-cell subcompartments, which is important for understanding chromosome spatial localization in single cells. Here we present scGHOST, a single-cell subcompartment annotation method using graph embedding with constrained random walk sampling. Applications of scGHOST to scHi-C data and contact maps derived from single-cell 3D genome imaging demonstrate reliable identification of single-cell subcompartments, offering insights into cell-to-cell variability of nuclear subcompartments. Using scHi-C data from complex tissues, scGHOST identifies cell-type-specific or allele-specific subcompartments linked to gene transcription across various cell types and developmental stages, suggesting functional implications of single-cell subcompartments. scGHOST is an effective method for annotating single-cell 3D genome subcompartments in a broad range of biological contexts.

MolFeSCue: enhancing molecular property prediction in data-limited and imbalanced contexts using few-shot and contrastive learning.

MOTIVATION: Predicting molecular properties is a pivotal task in various scientific domains, including drug discovery, material science, and computational chemistry. This problem is often hindered by the lack of annotated data and imbalanced class distributions, which pose significant challenges in developing accurate and robust predictive models. RESULTS: This study tackles these issues by employing pretrained molecular models within a few-shot learning framework. A novel dynamic contrastive loss function is utilized to further improve model performance in the situation of class imbalance. The proposed MolFeSCue framework not only facilitates rapid generalization from minimal samples, but also employs a contrastive loss function to extract meaningful molecular representations from imbalanced datasets. Extensive evaluations and comparisons of MolFeSCue and state-of-the-art algorithms have been conducted on multiple benchmark datasets, and the experimental data demonstrate our algorithm's effectiveness in molecular representations and its broad applicability across various pretrained models. Our findings underscore MolFeSCues potential to accelerate advancements in drug discovery. AVAILABILITY AND IMPLEMENTATION: We have made all the source code utilized in this study publicly accessible via GitHub at http://www.healthinformaticslab.org/supp/ or https://github.com/zhangruochi/MolFeSCue. The code (MolFeSCue-v1-00) is also available as the supplementary file of this paper.

PhyGCN: Pre-trained Hypergraph Convolutional Neural Networks with Self-supervised Learning.

Hypergraphs are powerful tools for modeling complex interactions across various domains, including biomedicine. However, learning meaningful node representations from hypergraphs remains a challenge. Existing supervised methods often lack generalizability, thereby limiting their real-world applications. We propose a new method, Pre-trained Hypergraph Convolutional Neural Networks with Self-supervised Learning (PhyGCN), which leverages hypergraph structure for self-supervision to enhance node representations. PhyGCN introduces a unique training strategy that integrates variable hyperedge sizes with self-supervised learning, enabling improved generalization to unseen data. Applications on multi-way chromatin interactions and polypharmacy side-effects demonstrate the effectiveness of PhyGCN. As a generic framework for high-order interaction datasets with abundant unlabeled data, PhyGCN holds strong potential for enhancing hypergraph node representations across various domains.

Concurrent profiling of multiscale 3D genome organization and gene expression in single mammalian cells.

The organization of mammalian genomes within the nucleus features a complex, multiscale three-dimensional (3D) architecture. The functional significance of these 3D genome features, however, remains largely elusive due to limited single-cell technologies that can concurrently profile genome organization and transcriptional activities. Here, we report GAGE-seq, a highly scalable, robust single-cell co-assay that simultaneously measures 3D genome structure and transcriptome within the same cell. Employing GAGE-seq on mouse brain cortex and human bone marrow CD34+ cells, we comprehensively characterized the intricate relationships between 3D genome and gene expression. We found that these multiscale 3D genome features collectively inform cell type-specific gene expressions, hence contributing to defining cell identity at the single-cell level. Integration of GAGE-seq data with spatial transcriptomic data revealed in situ variations of the 3D genome in mouse cortex. Moreover, our observations of lineage commitment in normal human hematopoiesis unveiled notable discordant changes between 3D genome organization and gene expression, underscoring a complex, temporal interplay at the single-cell level that is more nuanced than previously appreciated. Together, GAGE-seq provides a powerful, cost-effective approach for interrogating genome structure and gene expression relationships at the single-cell level across diverse biological contexts.

OCMR: A comprehensive framework for optical chemical molecular recognition.

Artificial intelligence (AI) has achieved significant progress in the field of drug discovery. AI-based tools have been used in all aspects of drug discovery, including chemical structure recognition. We propose a chemical structure recognition framework, Optical Chemical Molecular Recognition (OCMR), to improve the data extraction capability in practical scenarios compared with the rule-based and end-to-end deep learning models. The proposed OCMR framework enhances the recognition performances via the integration of local information in the topology of molecular graphs. OCMR handles complex tasks like non-canonical drawing and atomic group abbreviation and substantially improves the current state-of-the-art results on multiple public benchmark datasets and one internally curated dataset.

Orchestrating information across tissues via a novel multitask GAT framework to improve quantitative gene regulation relation modeling for survival analysis.

Survival analysis is critical to cancer prognosis estimation. High-throughput technologies facilitate the increase in the dimension of genic features, but the number of clinical samples in cohorts is relatively small due to various reasons, including difficulties in participant recruitment and high data-generation costs. Transcriptome is one of the most abundantly available OMIC (referring to the high-throughput data, including genomic, transcriptomic, proteomic and epigenomic) data types. This study introduced a multitask graph attention network (GAT) framework DQSurv for the survival analysis task. We first used a large dataset of healthy tissue samples to pretrain the GAT-based HealthModel for the quantitative measurement of the gene regulatory relations. The multitask survival analysis framework DQSurv used the idea of transfer learning to initiate the GAT model with the pretrained HealthModel and further fine-tuned this model using two tasks i.e. the main task of survival analysis and the auxiliary task of gene expression prediction. This refined GAT was denoted as DiseaseModel. We fused the original transcriptomic features with the difference vector between the latent features encoded by the HealthModel and DiseaseModel for the final task of survival analysis. The proposed DQSurv model stably outperformed the existing models for the survival analysis of 10 benchmark cancer types and an independent dataset. The ablation study also supported the necessity of the main modules. We released the codes and the pretrained HealthModel to facilitate the feature encodings and survival analysis of transcriptome-based future studies, especially on small datasets. The model and the code are available at http://www.healthinformaticslab.org/supp/.

Predicting the antigenic evolution of SARS-COV-2 with deep learning.

The relentless evolution of SARS-CoV-2 poses a significant threat to public health, as it adapts to immune pressure from vaccines and natural infections. Gaining insights into potential antigenic changes is critical but challenging due to the vast sequence space. Here, we introduce the Machine Learning-guided Antigenic Evolution Prediction (MLAEP), which combines structure modeling, multi-task learning, and genetic algorithms to predict the viral fitness landscape and explore antigenic evolution via in silico directed evolution. By analyzing existing SARS-CoV-2 variants, MLAEP accurately infers variant order along antigenic evolutionary trajectories, correlating with corresponding sampling time. Our approach identified novel mutations in immunocompromised COVID-19 patients and emerging variants like XBB1.5. Additionally, MLAEP predictions were validated through in vitro neutralizing antibody binding assays, demonstrating that the predicted variants exhibited enhanced immune evasion. By profiling existing variants and predicting potential antigenic changes, MLAEP aids in vaccine development and enhances preparedness against future SARS-CoV-2 variants.

AB-Gen: Antibody Library Design with Generative Pre-trained Transformer and Deep Reinforcement Learning.

Antibody leads must fulfill multiple desirable properties to be clinical candidates. Primarily due to the low throughput in the experimental procedure, the need for such multi-property optimization causes the bottleneck in preclinical antibody discovery and development, because addressing one issue usually causes another. We developed a reinforcement learning (RL) method, named AB-Gen, for antibody library design using a generative pre-trained transformer (GPT) as the policy network of the RL agent. We showed that this model can learn the antibody space of heavy chain complementarity determining region 3 (CDRH3) and generate sequences with similar property distributions. Besides, when using human epidermal growth factor receptor-2 (HER2) as the target, the agent model of AB-Gen was able to generate novel CDRH3 sequences that fulfill multi-property constraints. Totally, 509 generated sequences were able to pass all property filters, and three highly conserved residues were identified. The importance of these residues was further demonstrated by molecular dynamics simulations, consolidating that the agent model was capable of grasping important information in this complex optimization task. Overall, the AB-Gen method is able to design novel antibody sequences with an improved success rate than the traditional propose-then-filter approach. It has the potential to be used in practical antibody design, thus empowering the antibody discovery and development process. The source code of AB-Gen is freely available at Zenodo (https://doi.org/10.5281/zenodo.7657016) and BioCode (https://ngdc.cncb.ac.cn/biocode/tools/BT007341).

scGHOST: Identifying single-cell 3D genome subcompartments.

New single-cell Hi-C (scHi-C) technologies enable probing of the genome-wide cell-to-cell variability in 3D genome organization from individual cells. Several computational methods have been developed to reveal single-cell 3D genome features based on scHi-C data, including A/B compartments, topologically-associating domains, and chromatin loops. However, no scHi-C analysis method currently exists for annotating single-cell subcompartments, which are crucial for providing a more refined view of large-scale chromosome spatial localization in single cells. Here, we present scGhost, a single-cell subcompartment annotation method based on graph embedding with constrained random walk sampling. Applications of scGhost to scHi-C data and single-cell 3D genome imaging data demonstrate the reliable identification of single-cell subcompartments and offer new insights into cell-to-cell variability of nuclear subcompartments. Using scHi-C data from the human prefrontal cortex, scGhost identifies cell type-specific subcompartments that are strongly connected to cell type-specific gene expression, suggesting the functional implications of single-cell subcompartments. Overall, scGhost is an effective new method for single-cell 3D genome subcompartment annotation based on scHi-C data for a broad range of biological contexts.

Single-cell multi-scale footprinting reveals the modular organization of DNA regulatory elements.

Cis-regulatory elements control gene expression and are dynamic in their structure, reflecting changes to the composition of diverse effector proteins over time1-3. Here we sought to connect the structural changes at cis-regulatory elements to alterations in cellular fate and function. To do this we developed PRINT, a computational method that uses deep learning to correct sequence bias in chromatin accessibility data and identifies multi-scale footprints of DNA-protein interactions. We find that multi-scale footprints enable more accurate inference of TF and nucleosome binding. Using PRINT with single-cell multi-omics, we discover wide-spread changes to the structure and function of candidate cis-regulatory elements (cCREs) across hematopoiesis, wherein nucleosomes slide, expose DNA for TF binding, and promote gene expression. Activity segmentation using the co-variance across cell states identifies "sub-cCREs" as modular cCRE subunits of regulatory DNA. We apply this single-cell and PRINT approach to characterize the age-associated alterations to cCREs within hematopoietic stem cells (HSCs). Remarkably, we find a spectrum of aging alterations among HSCs corresponding to a global gain of sub-cCRE activity while preserving cCRE accessibility. Collectively, we reveal the functional importance of cCRE structure across cell states, highlighting changes to gene regulation at single-cell and single-base-pair resolution.

A comprehensive benchmarking with practical guidelines for cellular deconvolution of spatial transcriptomics.

Spatial transcriptomics technologies are used to profile transcriptomes while preserving spatial information, which enables high-resolution characterization of transcriptional patterns and reconstruction of tissue architecture. Due to the existence of low-resolution spots in recent spatial transcriptomics technologies, uncovering cellular heterogeneity is crucial for disentangling the spatial patterns of cell types, and many related methods have been proposed. Here, we benchmark 18 existing methods resolving a cellular deconvolution task with 50 real-world and simulated datasets by evaluating the accuracy, robustness, and usability of the methods. We compare these methods comprehensively using different metrics, resolutions, spatial transcriptomics technologies, spot numbers, and gene numbers. In terms of performance, CARD, Cell2location, and Tangram are the best methods for conducting the cellular deconvolution task. To refine our comparative results, we provide decision-tree-style guidelines and recommendations for method selection and their additional features, which will help users easily choose the best method for fulfilling their concerns.

Optimization of binding affinities in chemical space with generative pre-trained transformer and deep reinforcement learning.

Background: The key challenge in drug discovery is to discover novel compounds with desirable properties. Among the properties, binding affinity to a target is one of the prerequisites and usually evaluated by molecular docking or quantitative structure activity relationship (QSAR) models. Methods: In this study, we developed SGPT-RL, which uses a generative pre-trained transformer (GPT) as the policy network of the reinforcement learning (RL) agent to optimize the binding affinity to a target. SGPT-RL was evaluated on the Moses distribution learning benchmark and two goal-directed generation tasks, with Dopamine Receptor D2 (DRD2) and Angiotensin-Converting Enzyme 2 (ACE2) as the targets. Both QSAR model and molecular docking were implemented as the optimization goals in the tasks. The popular Reinvent method was used as the baseline for comparison. Results: The results on the Moses benchmark showed that SGPT-RL learned good property distributions and generated molecules with high validity and novelty. On the two goal-directed generation tasks, both SGPT-RL and Reinvent were able to generate valid molecules with improved target scores. The SGPT-RL method achieved better results than Reinvent on the ACE2 task, where molecular docking was used as the optimization goal. Further analysis shows that SGPT-RL learned conserved scaffold patterns during exploration. Conclusions: The superior performance of SGPT-RL in the ACE2 task indicates that it can be applied to the virtual screening process where molecular docking is widely used as the criteria. Besides, the scaffold patterns learned by SGPT-RL during the exploration process can assist chemists to better design and discover novel lead candidates.

Ultrafast and interpretable single-cell 3D genome analysis with Fast-Higashi.

Single-cell Hi-C (scHi-C) technologies can probe three-dimensional (3D) genome structures in individual cells. However, existing scHi-C analysis methods are hindered by the data quality and complex 3D genome patterns. The lack of computational scalability and interpretability poses further challenges for large-scale analysis. Here, we introduce Fast-Higashi, an ultrafast and interpretable method based on tensor decomposition and partial random walk with restart, enabling joint identification of cell identities and chromatin meta-interactions from sparse scHi-C data. Extensive evaluations demonstrate the advantage of Fast-Higashi over existing methods, leading to improved delineation of rare cell types and continuous developmental trajectories. Fast-Higashi can directly identify 3D genome features that define distinct cell types and help elucidate cell-type-specific connections between genome structure and function. Moreover, Fast-Higashi can generalize to incorporate other single-cell omics data. Fast-Higashi provides a highly efficient and interpretable scHi-C analysis solution that is applicable to a broad range of biological contexts.

A polygenic stacking classifier revealed the complicated platelet transcriptomic landscape of adult immune thrombocytopenia.

Immune thrombocytopenia (ITP) is an autoimmune disease with the typical symptom of a low platelet count in blood. ITP demonstrated age and sex biases in both occurrences and prognosis, and adult ITP was mainly induced by the living environments. The current diagnosis guideline lacks the integration of molecular heterogenicity. This study recruited the largest cohort of platelet transcriptome samples. A comprehensive procedure of feature selection, feature engineering, and stacking classification was carried out to detect the ITP biomarkers using RNA sequencing (RNA-seq) transcriptomes. The 40 detected biomarkers were loaded to train the final ITP detection model, with an overall accuracy 0.974. The biomarkers suggested that ITP onset may be associated with various transcribed components, including protein-coding genes, long intergenic non-coding RNA (lincRNA) genes, and pseudogenes with apparent transcriptions. The delivered ITP detection model may also be utilized as a complementary ITP diagnosis tool. The code and the example dataset is freely available on http://www.healthinformaticslab.org/supp/resources.php.

Describe Molecules by a Heterogeneous Graph Neural Network with Transformer-like Attention for Supervised Property Predictions.

Machine learning and deep learning have facilitated various successful studies of molecular property predictions. The rapid development of natural language processing and graph neural network (GNN) further pushed the state-of-the-art prediction performance of molecular property to a new level. A geometric graph could describe a molecular structure with atoms as the nodes and bonds as the edges. Therefore, a graph neural network may be trained to better represent a molecular structure. The existing GNNs assumed homogeneous types of atoms and bonds, which may miss important information between different types of atoms or bonds. This study represented a molecule using a heterogeneous graph neural network (MolHGT), in which there were different types of nodes and different types of edges. A transformer reading function of virtual nodes was proposed to aggregate all the nodes, and a molecule graph may be represented from the hidden states of the virtual nodes. This proof-of-principle study demonstrated that the proposed MolHGT network improved the existing studies of molecular property predictions. The source code and the training/validation/test splitting details are available at https://github.com/zhangruochi/Mol-HGT.

Multiscale and integrative single-cell Hi-C analysis with Higashi.

Single-cell Hi-C (scHi-C) can identify cell-to-cell variability of three-dimensional (3D) chromatin organization, but the sparseness of measured interactions poses an analysis challenge. Here we report Higashi, an algorithm based on hypergraph representation learning that can incorporate the latent correlations among single cells to enhance overall imputation of contact maps. Higashi outperforms existing methods for embedding and imputation of scHi-C data and is able to identify multiscale 3D genome features in single cells, such as compartmentalization and TAD-like domain boundaries, allowing refined delineation of their cell-to-cell variability. Moreover, Higashi can incorporate epigenomic signals jointly profiled in the same cell into the hypergraph representation learning framework, as compared to separate analysis of two modalities, leading to improved embeddings for single-nucleus methyl-3C data. In an scHi-C dataset from human prefrontal cortex, Higashi identifies connections between 3D genome features and cell-type-specific gene regulation. Higashi can also potentially be extended to analyze single-cell multiway chromatin interactions and other multimodal single-cell omics data.

The 3D Genome Structure of Single Cells.

The spatial organization of the genome in the cell nucleus is pivotal to cell function. However, how the 3D genome organization and its dynamics influence cellular phenotypes remains poorly understood. The very recent development of single-cell technologies for probing the 3D genome, especially single-cell Hi-C (scHi-C), has ushered in a new era of unveiling cell-to-cell variability of 3D genome features at an unprecedented resolution. Here, we review recent developments in computational approaches to the analysis of scHi-C, including data processing, dimensionality reduction, imputation for enhancing data quality, and the revealing of 3D genome features at single-cell resolution. While much progress has been made in computational method development to analyze single-cell 3D genomes, substantial future work is needed to improve data interpretation and multimodal data integration, which are critical to reveal fundamental connections between genome structure and function among heterogeneous cell populations in various biological contexts.