Unveil cis-acting combinatorial mRNA motifs by interpreting deep neural network.

SUMMARY: Cis-acting mRNA elements play a key role in the regulation of mRNA stability and translation efficiency. Revealing the interactions of these elements and their impact plays a crucial role in understanding the regulation of the mRNA translation process, which supports the development of mRNA-based medicine or vaccines. Deep neural networks (DNN) can learn complex cis-regulatory codes from RNA sequences. However, extracting these cis-regulatory codes efficiently from DNN remains a significant challenge. Here, we propose a method based on our toolkit NeuronMotif and motif mutagenesis, which not only enables the discovery of diverse and high-quality motifs but also efficiently reveals motif interactions. By interpreting deep-learning models, we have discovered several crucial motifs that impact mRNA translation efficiency and stability, as well as some unknown motifs or motif syntax, offering novel insights for biologists. Furthermore, we note that it is challenging to enrich motif syntax in datasets composed of randomly generated sequences, and they may not contain sufficient biological signals. AVAILABILITY AND IMPLEMENTATION: The source code and data used to produce the results and analyses presented in this manuscript are available from GitHub (https://github.com/WangLabTHU/combmotif).

GPro: generative AI-empowered toolkit for promoter design.

MOTIVATION: Promoters with desirable properties are crucial in biotechnological applications. Generative AI (GenAI) has demonstrated potential in creating novel synthetic promoters with significantly enhanced functionality. However, these methods' reliance on various programming frameworks and specific task-oriented contexts limits their flexibilities. Overcoming these limitations is essential for researchers to fully leverage the power of GenAI to design promoters for their tasks. RESULTS: Here, we introduce GPro (Generative AI-empowered toolkit for promoter design), a user-friendly toolkit that integrates a collection of cutting-edge GenAI-empowered approaches for promoter design. This toolkit provides a standardized pipeline covering essential promoter design processes, including training, optimization, and evaluation. Several detailed demos are provided to reproduce state-of-the-art promoter design pipelines. GPro's user-friendly interface makes it accessible to a wide range of users including non-AI experts. It also offers a variety of optional algorithms for each design process, and gives users the flexibility to compare methods and create customized pipelines. AVAILABILITY AND IMPLEMENTATION: GPro is released as an open-source software under the MIT license. The source code for GPro is available on GitHub for Linux, macOS, and Windows: https://github.com/WangLabTHU/GPro, and is available for download via Zenodo repository at https://zenodo.org/doi/10.5281/zenodo.10681733.

DIProT: A deep learning based interactive toolkit for efficient and effective Protein design.

The protein inverse folding problem, designing amino acid sequences that fold into desired protein structures, is a critical challenge in biological sciences. Despite numerous data-driven and knowledge-driven methods, there remains a need for a user-friendly toolkit that effectively integrates these approaches for in-silico protein design. In this paper, we present DIProT, an interactive protein design toolkit. DIProT leverages a non-autoregressive deep generative model to solve the inverse folding problem, combined with a protein structure prediction model. This integration allows users to incorporate prior knowledge into the design process, evaluate designs in silico, and form a virtual design loop with human feedback. Our inverse folding model demonstrates competitive performance in terms of effectiveness and efficiency on TS50 and CATH4.2 datasets, with promising sequence recovery and inference time. Case studies further illustrate how DIProT can facilitate user-guided protein design.

Deep flanking sequence engineering for efficient promoter design using DeepSEED.

Designing promoters with desirable properties is essential in synthetic biology. Human experts are skilled at identifying strong explicit patterns in small samples, while deep learning models excel at detecting implicit weak patterns in large datasets. Biologists have described the sequence patterns of promoters via transcription factor binding sites (TFBSs). However, the flanking sequences of cis-regulatory elements, have long been overlooked and often arbitrarily decided in promoter design. To address this limitation, we introduce DeepSEED, an AI-aided framework that efficiently designs synthetic promoters by combining expert knowledge with deep learning techniques. DeepSEED has demonstrated success in improving the properties of Escherichia coli constitutive, IPTG-inducible, and mammalian cell doxycycline (Dox)-inducible promoters. Furthermore, our results show that DeepSEED captures the implicit features in flanking sequences, such as k-mer frequencies and DNA shape features, which are crucial for determining promoter properties.

NeuronMotif: Deciphering cis-regulatory codes by layer-wise demixing of deep neural networks.

Discovering DNA regulatory sequence motifs and their relative positions is vital to understanding the mechanisms of gene expression regulation. Although deep convolutional neural networks (CNNs) have achieved great success in predicting cis-regulatory elements, the discovery of motifs and their combinatorial patterns from these CNN models has remained difficult. We show that the main difficulty is due to the problem of multifaceted neurons which respond to multiple types of sequence patterns. Since existing interpretation methods were mainly designed to visualize the class of sequences that can activate the neuron, the resulting visualization will correspond to a mixture of patterns. Such a mixture is usually difficult to interpret without resolving the mixed patterns. We propose the NeuronMotif algorithm to interpret such neurons. Given any convolutional neuron (CN) in the network, NeuronMotif first generates a large sample of sequences capable of activating the CN, which typically consists of a mixture of patterns. Then, the sequences are "demixed" in a layer-wise manner by backward clustering of the feature maps of the involved convolutional layers. NeuronMotif can output the sequence motifs, and the syntax rules governing their combinations are depicted by position weight matrices organized in tree structures. Compared to existing methods, the motifs found by NeuronMotif have more matches to known motifs in the JASPAR database. The higher-order patterns uncovered for deep CNs are supported by the literature and ATAC-seq footprinting. Overall, NeuronMotif enables the deciphering of cis-regulatory codes from deep CNs and enhances the utility of CNN in genome interpretation.

Enabling technology and core theory of synthetic biology.

Synthetic biology provides a new paradigm for life science research ("build to learn") and opens the future journey of biotechnology ("build to use"). Here, we discuss advances of various principles and technologies in the mainstream of the enabling technology of synthetic biology, including synthesis and assembly of a genome, DNA storage, gene editing, molecular evolution and de novo design of function proteins, cell and gene circuit engineering, cell-free synthetic biology, artificial intelligence (AI)-aided synthetic biology, as well as biofoundries. We also introduce the concept of quantitative synthetic biology, which is guiding synthetic biology towards increased accuracy and predictability or the real rational design. We conclude that synthetic biology will establish its disciplinary system with the iterative development of enabling technologies and the maturity of the core theory.

Evaluating methylation of human ribosomal DNA at each CpG site reveals its utility for cancer detection using cell-free DNA.

Ribosomal deoxyribonucleic acid (DNA) (rDNA) repeats are tandemly located on five acrocentric chromosomes with up to hundreds of copies in the human genome. DNA methylation, the most well-studied epigenetic mechanism, has been characterized for most genomic regions across various biological contexts. However, rDNA methylation patterns remain largely unexplored due to the repetitive structure. In this study, we designed a specific mapping strategy to investigate rDNA methylation patterns at each CpG site across various physiological and pathological processes. We found that CpG sites on rDNA could be categorized into two types. One is within or adjacent to transcribed regions; the other is distal to transcribed regions. The former shows highly variable methylation levels across samples, while the latter shows stable high methylation levels in normal tissues but severe hypomethylation in tumors. We further showed that rDNA methylation profiles in plasma cell-free DNA could be used as a biomarker for cancer detection. It shows good performances on public datasets, including colorectal cancer [area under the curve (AUC) = 0.85], lung cancer (AUC = 0.84), hepatocellular carcinoma (AUC = 0.91) and in-house generated hepatocellular carcinoma dataset (AUC = 0.96) even at low genome coverage (<1×). Taken together, these findings broaden our understanding of rDNA regulation and suggest the potential utility of rDNA methylation features as disease biomarkers.

MeConcord: a new metric to quantitatively characterize DNA methylation heterogeneity across reads and CpG sites.

MOTIVATION: Intermediately methylated regions occupy a significant fraction of the human genome and are closely associated with epigenetic regulations or cell-type deconvolution of bulk data. However, these regions show distinct methylation patterns, corresponding to different biological mechanisms. Although there have been some metrics developed for investigating these regions, the high noise sensitivity limits the utility for distinguishing distinct methylation patterns. RESULTS: We proposed a method named MeConcord to measure local methylation concordance across reads and CpG sites, respectively. MeConcord showed the most stable performance in distinguishing distinct methylation patterns ('identical', 'uniform' and 'disordered') compared with other metrics. Applying MeConcord to the whole genome data across 25 cell lines or primary cells or tissues, we found that distinct methylation patterns were associated with different genomic characteristics, such as CTCF binding or imprinted genes. Further, we showed the differences of CpG island hypermethylation patterns between senescence and tumorigenesis by using MeConcord. MeConcord is a powerful method to study local read-level methylation patterns for both the whole genome and specific regions of interest. AVAILABILITY AND IMPLEMENTATION: MeConcord is available at https://github.com/WangLabTHU/MeConcord. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

DeSP: a systematic DNA storage error simulation pipeline.

BACKGROUND: Using DNA as a storage medium is appealing due to the information density and longevity of DNA, especially in the era of data explosion. A significant challenge in the DNA data storage area is to deal with the noises introduced in the channel and control the trade-off between the redundancy of error correction codes and the information storage density. As running DNA data storage experiments in vitro is still expensive and time-consuming, a simulation model is needed to systematically optimize the redundancy to combat the channel's particular noise structure. RESULTS: Here, we present DeSP, a systematic DNA storage error Simulation Pipeline, which simulates the errors generated from all DNA storage stages and systematically guides the optimization of encoding redundancy. It covers both the sequence lost and the within-sequence errors in the particular context of the data storage channel. With this model, we explained how errors are generated and passed through different stages to form final sequencing results, analyzed the influence of error rate and sampling depth to final error rates, and demonstrated how to systemically optimize redundancy design in silico with the simulation model. These error simulation results are consistent with the in vitro experiments. CONCLUSIONS: DeSP implemented in Python is freely available on Github ( https://github.com/WangLabTHU/DeSP ). It is a flexible framework for systematic error simulation in DNA storage and can be adapted to a wide range of experiment pipelines.

hECA: The cell-centric assembly of a cell atlas.

The accumulation of massive single-cell omics data provides growing resources for building biomolecular atlases of all cells of human organs or the whole body. The true assembly of a cell atlas should be cell-centric rather than file-centric. We developed a unified informatics framework for seamless cell-centric data assembly and built the human Ensemble Cell Atlas (hECA) from scattered data. hECA v1.0 assembled 1,093,299 labeled human cells from 116 published datasets, covering 38 organs and 11 systems. We invented three new methods of atlas applications based on the cell-centric assembly: "in data" cell sorting for targeted data retrieval with customizable logic expressions, "quantitative portraiture" for multi-view representations of biological entities, and customizable reference creation for generating references for automatic annotations. Case studies on agile construction of user-defined sub-atlases and "in data" investigation of CAR-T off-targets in multiple organs showed the great potential enabled by the cell-centric ensemble atlas.

In silico and in vitro protocols for quantifying gene expression noise modulated by microRNAs.

Characterizing the noise modulation pattern of microRNA is valuable for both microRNA function analysis and synthetic biology applications. Here we propose a coarse-grained model to simulate how the properties of microRNAs, competing RNAs, and microRNA response elements affect gene expression noise. We also detail an experimental protocol based on synthetic gene circuits and flow cytometry to quantify the noise. This framework is easy-to-use for the study and application of both microRNA and gene expression noise. For complete details on the use and execution of this protocol, please refer to Wei et al. (2021).

Highly Regional Genes: graph-based gene selection for single-cell RNA-seq data.

Gene selection is an indispensable step for analyzing noisy and high-dimensional single-cell RNA-seq (scRNA-seq) data. Compared with the commonly used variance-based methods, by mimicking the human maker selection in the 2D visualization of cells, a new feature selection method called HRG (Highly Regional Genes) is proposed to find the informative genes, which show regional expression patterns in the cell-cell similarity network. We mathematically find the optimal expression patterns that can maximize the proposed scoring function. In comparison with several unsupervised methods, HRG shows high accuracy and robustness, and can increase the performance of downstream cell clustering and gene correlation analysis. Also, it is applicable for selecting informative genes of sequencing-based spatial transcriptomic data.

YTHDF3 modulates hematopoietic stem cells by recognizing RNA m6A modification on Ccnd1.

Hematopoietic stem cells (HSC) give rise to the cells of the blood system over the whole lifespan. N6-methyladenosine (m6A), the most prevalent RNA modification, modulates gene expression via the processes of "writing" and "reading". Recent studies showed that m6A "writer" genes (Mettl3 and Mettl14) play an essential role in HSC. However, which reader deciphers the m6A modification to modulate HSC remains unknown. In this study, we observed that dysfunction of Ythdf3 and Ccnd1 severely impaired the reconstitution capacity of HSC, which phenocopies Mettl3-deficient HSC. Dysfunction of Ythdf3 and Mettl3 results in a translational defect of Ccnd1. Ythdf3 and Mettl3 regulate HSC by transmitting m6A RNA methylation on the 5' untranslated region of Ccnd1. Enforced Ccnd1 expression completely rescued the defect of Ythdf3-/- HSC and partially rescued Mettl3-compromised HSC. Taken together, this study identified, for the first time, that Ccnd1 is the target of METTL3 and YTHDF3 to transmit the m6A RNA methylation signal and thereby regulate the reconstitution capacity of HSC.

ARIC: accurate and robust inference of cell type proportions from bulk gene expression or DNA methylation data.

Quantifying cell proportions, especially for rare cell types in some scenarios, is of great value in tracking signals associated with certain phenotypes or diseases. Although some methods have been proposed to infer cell proportions from multicomponent bulk data, they are substantially less effective for estimating the proportions of rare cell types which are highly sensitive to feature outliers and collinearity. Here we proposed a new deconvolution algorithm named ARIC to estimate cell type proportions from gene expression or DNA methylation data. ARIC employs a novel two-step marker selection strategy, including collinear feature elimination based on the component-wise condition number and adaptive removal of outlier markers. This strategy can systematically obtain effective markers for weighted $\upsilon$-support vector regression to ensure a robust and precise rare proportion prediction. We showed that ARIC can accurately estimate fractions in both DNA methylation and gene expression data from different experiments. We further applied ARIC to the survival prediction of ovarian cancer and the condition monitoring of chronic kidney disease, and the results demonstrate the high accuracy and robustness as well as clinical potentials of ARIC. Taken together, ARIC is a promising tool to solve the deconvolution problem of bulk data where rare components are of vital importance.

Single-cell Transcriptomes Reveal Characteristics of MicroRNAs in Gene Expression Noise Reduction.

Isogenic cells growing in identical environments show cell-to-cell variations because of the stochasticity in gene expression. High levels of variation or noise can disrupt robust gene expression and result in tremendous consequences for cell behaviors. In this work, we showed evidence from single-cell RNA sequencing data analysis that microRNAs (miRNAs) can reduce gene expression noise at the mRNA level in mouse cells. We identified that the miRNA expression level, number of targets, target pool abundance, and miRNA-target interaction strength are the key features contributing to noise repression. miRNAs tend to work together in cooperative subnetworks to repress target noise synergistically in a cell type-specific manner. By building a physical model of post-transcriptional regulation and observing in synthetic gene circuits, we demonstrated that accelerated degradation with elevated transcriptional activation of the miRNA target provides resistance to extrinsic fluctuations. Together, through the integrated analysis of single-cell RNA and miRNA expression profiles, we demonstrated that miRNAs are important post-transcriptional regulators for reducing gene expression noise and conferring robustness to biological processes.

Characterizing microRNA-mediated modulation of gene expression noise and its effect on synthetic gene circuits.

MicroRNAs (miRNAs) have been shown to modulate gene expression noise, but less is known about how miRNAs with different properties may regulate noise differently. Here, we investigate the role of competing RNAs and the composition of miRNA response elements (MREs) in modulating noise. We find that weak competing RNAs could introduce lower noise than strong competing RNAs. In comparison with a single MRE, both repetitive and composite MREs can reduce the noise at low expression, but repetitive MREs can elevate the noise remarkably at high expression. We further observed the behavior of a synthetic cell-type classifier with miRNAs as inputs and find that miRNAs and MREs that could introduce higher noise tend to enhance cell state transition. These results provide a systematic and quantitative understanding of the function of miRNAs in controlling gene expression noise and the utilization of miRNAs to modulate the behavior of synthetic gene circuits.

DISMIR: Deep learning-based noninvasive cancer detection by integrating DNA sequence and methylation information of individual cell-free DNA reads.

Detecting cancer signals in cell-free DNA (cfDNA) high-throughput sequencing data is emerging as a novel noninvasive cancer detection method. Due to the high cost of sequencing, it is crucial to make robust and precise predictions with low-depth cfDNA sequencing data. Here we propose a novel approach named DISMIR, which can provide ultrasensitive and robust cancer detection by integrating DNA sequence and methylation information in plasma cfDNA whole-genome bisulfite sequencing (WGBS) data. DISMIR introduces a new feature termed as 'switching region' to define cancer-specific differentially methylated regions, which can enrich the cancer-related signal at read-resolution. DISMIR applies a deep learning model to predict the source of every single read based on its DNA sequence and methylation state and then predicts the risk that the plasma donor is suffering from cancer. DISMIR exhibited high accuracy and robustness on hepatocellular carcinoma detection by plasma cfDNA WGBS data even at ultralow sequencing depths. Further analysis showed that DISMIR tends to be insensitive to alterations of single CpG sites' methylation states, which suggests DISMIR could resist to technical noise of WGBS. All these results showed DISMIR with the potential to be a precise and robust method for low-cost early cancer detection.

The loss of heterochromatin is associated with multiscale three-dimensional genome reorganization and aberrant transcription during cellular senescence.

Heterochromatin remodeling is critical for various cell processes. In particular, the "loss of heterochromatin" phenotype in cellular senescence is associated with the process of aging and age-related disorders. Although biological processes of senescent cells, including senescence-associated heterochromatin foci (SAHF) formation, chromosome compaction, and redistribution of key proteins, have been closely associated with high-order chromatin structure, the relationship between the high-order chromatin reorganization and the loss of heterochromatin phenotype during senescence has not been fully understood. By using senescent and deep senescent fibroblasts induced by DNA damage harboring the "loss of heterochromatin" phenotype, we observed progressive 3D reorganization of heterochromatin during senescence. Facultative and constitutive heterochromatin marked by H3K27me3 and H3K9me3, respectively, show different alterations. Facultative heterochromatin tends to switch from the repressive B-compartment to the active A-compartment, whereas constitutive heterochromatin shows no significant changes at the compartment level but enhanced interactions between themselves. Both types of heterochromatin show increased chromatin accessibility and gene expression leakage during senescence. Furthermore, increased chromatin accessibility in potential CTCF binding sites accompanies the establishment of novel loops in constitutive heterochromatin. Finally, we also observed aberrant expression of repetitive elements, including LTR (long terminal repeat) and satellite classes. Overall, facultative and constitutive heterochromatin show both similar and distinct multiscale alterations in the 3D map, chromatin accessibility, and gene expression leakage. This study provides an epigenomic map of heterochromatin reorganization during senescence.

cfDNApipe: a comprehensive quality control and analysis pipeline for cell-free DNA high-throughput sequencing data.

MOTIVATION: Cell-free DNA (cfDNA) is gaining substantial attention from both biological and clinical fields as a promising marker for liquid biopsy. Many aspects of disease-related features have been discovered from cfDNA high-throughput sequencing (HTS) data. However, there is still a lack of integrative and systematic tools for cfDNA HTS data analysis and quality control (QC). RESULTS: Here, we propose cfDNApipe, an easy-to-use and systematic python package for cfDNA whole-genome sequencing (WGS) and whole-genome bisulfite sequencing (WGBS) data analysis. It covers the entire analysis pipeline for the cfDNA data, including raw sequencing data processing, QC and sophisticated statistical analysis such as detecting copy number variations (CNVs), differentially methylated regions and DNA fragment size alterations. cfDNApipe provides one-command-line-execution pipelines and flexible application programming interfaces for customized analysis. AVAILABILITY AND IMPLEMENTATION: https://xwanglabthu.github.io/cfDNApipe/. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.