PaperR-(4)-2022-10

我们在这个 Paper Reading 系列分享一些和生物医学相关的文献。

以下是我们十月份看到的一些比较有意思的文章：

生物医学研究

组学数据分析

Nature Inferring and perturbing cell fate regulomes in human brain organoids

Highlights: "Here we acquire single-cell transcriptome and accessible chromatin data over a dense time course in human organoids covering neuroepithelial formation, patterning, brain regionalization and neurogenesis, and identify temporally dynamic and brain-region-specific regulatory regions."

Multi-omic atlas of brain organoid development reveals developmental hierarchies and critical stages of fate decision.

Link: https://www.nature.com/articles/s41586-022-05279-8

Nature The co-evolution of the genome and epigenome in colorectal cancer

Highlights: "DNA mutations alone do not fully explain malignant transformation. Here we investigate the co-evolution of the genome and epigenome of colorectal tumours at single-clone resolution using spatial multi-omic profiling of individual glands. Genome-wide alterations in accessibility for transcription factor binding involved CTCF, downregulation of interferon and increased accessibility for SOX and HOX transcription factor families, suggesting the involvement of developmental genes during tumourigenesis. Somatic chromatin accessibility alterations were heritable and distinguished adenomas from cancers. Mutational signature analysis showed that the epigenome in turn influences the accumulation of DNA mutations. This study provides a map of genetic and epigenetic tumour heterogeneity, with fundamental implications for understanding colorectal cancer biology."

Spatial single-gland multi-omics.

Link: https://www.nature.com/articles/s41586-022-05202-1

Nature Single-cell genomic variation induced by mutational processes in cancer

Highlights: "Here, by applying scaled single-cell whole-genome sequencing to wild-type, TP53-deficient and TP53-deficient;BRCA1-deficient or TP53-deficient;BRCA2-deficient mammary epithelial cells (13,818 genomes), and to primary TNBC and HGSC cells (22,057 genomes), we identify three distinct ‘foreground’ mutational patterns that are defined by cell-to-cell structural variation. In TNBC and HGSC, clone-specific high-level amplifications in known oncogenes were highly prevalent in tumours bearing fold-back inversions, relative to tumours with homologous recombination deficiency, and were associated with increased clone-to-clone phenotypic variation."

Single-cell genome properties of CRISPR–Cas9-derived isogenic genotypes of 184-hTERT mammary epithelial cell lines.

Link: https://www.nature.com/articles/s41586-022-05249-0

Nature Phenotypic plasticity and genetic control in colorectal cancer evolution

Highlights: "Here we show that intratumour genetic ancestry only infrequently affects gene expression traits and subclonal evolution in colorectal cancer (CRC). Using spatially resolved paired whole-genome and transcriptome sequencing, we find that the majority of intratumour variation in gene expression is not strongly heritable but rather ‘plastic’. Together, our data suggest that most genetic intratumour variation in CRC has no major phenotypic consequence and that transcriptional plasticity is, instead, widespread within a tumour."

Link: https://www.nature.com/articles/s41586-022-05311-x

其它关注

Cell Spatially resolved epigenomic profiling of single cells in complex tissues

Highlights: "Here, we report a method for spatially resolved epigenomic profiling of single cells using in situ tagmentation and transcription followed by multiplexed imaging. We demonstrated the ability to profile histone modifications marking active promoters, putative enhancers, and silent promoters in individual cells."

Epigenomic MERFISH

Link: https://www.cell.com/cell/fulltext/S0092-8674(22)01254-5

生物信息方法

CNV 检测

bioRxiv Refphase: Multi-sample reference phasing reveals haplotype-specific copy number heterogeneity

Highlights: "The sequencing of multiple tumour samples from a patient’s disease is an increasingly common practice. We introduce Refphase, an algorithm that leverages this multi-sampling approach to infer haplotype-specific copy numbers through multi-sample reference phasing."

Overview of Refphase algorithm.

Link: https://www.biorxiv.org/content/10.1101/2022.10.13.511885v1.full

RNA Velocity

Nature Biotechnology Multi-omic single-cell velocity models epigenome–transcriptome interactions and improves cell fate prediction

Highlights: "We developed MultiVelo, a differential equation model of gene expression that extends the RNA velocity framework to incorporate epigenomic data. MultiVelo uses a probabilistic latent variable model to estimate the switch time and rate parameters of chromatin accessibility and gene expression and improves the accuracy of cell fate prediction compared to velocity estimates from RNA only. We also find four types of cell states: two states in which epigenome and transcriptome are coupled and two distinct decoupled states."

Schematic of MultiVelo approach.

Link: https://www.nature.com/articles/s41587-022-01476-y

bioRxiv Towards Hierarchical Causal Representation Learning for Nonstationary Multi-Omics Data

Highlights: "Gene expression responses could lag behind changes in chromatin accessibility. We propose HALO, which factorizes these two modalities, scATAC- and scRNA-seq data, into both coupled (changing dependently) and decoupled (changing independently) latent representations, allowing us to identify the dynamic interplay between chromatin accessibility and transcription through temporal modulations."

HALO models coupling and decoupling with changing causal modules.

Link: https://www.biorxiv.org/content/10.1101/2022.10.17.512602v2.full

bioRxiv Unraveling causal gene regulation from the RNA velocity graph using Velorama

Highlights: "Current Gene regulatory network (GRN) inference methods require a total ordering of cells along a linear pseudotemporal axis, which is biologically inappropriate since trajectories with branches cannot be reduced to a single time axis. Here, we introduce Velorama, which represents scRNA-seq differentiation dynamics as a partial ordering of cells and operates on the directed acyclic graph (DAG) of cells constructed from pseudotime or RNA velocity measurements. To our knowledge, Velorama is the first GRN inference method that can work directly with RNA velocity-based cell-to-cell transition probabilities."

Schematic of Velorama

Link: https://www.biorxiv.org/content/10.1101/2022.10.18.512766v1.full

bioRxiv Modeling Single-Cell Dynamics Using Unbalanced Parameterized Monge Maps

Highlights: "Optimal Transport (OT) has proven useful to infer single-cell trajectories of developing biological systems by aligning distributions across time points. Recently, Parameterized Monge Maps (PMM) were introduced to learn the optimal map between two distributions. Here, we apply PMM to model single-cell dynamics and show that PMM fails to account for asymmetric shifts in cell state distributions. To alleviate this limitation, we propose Unbalanced Parameterised Monge Maps (UPMM). We first describe the novel formulation and show on synthetic data how our method extends discrete unbalanced OT to the continuous domain. Then, we demonstrate that UPMM outperforms well-established trajectory inference methods on real-world developmental single-cell data."

Different maps on data drawn from a mixture of uniform distribution

Link: https://www.biorxiv.org/content/10.1101/2022.10.04.510766v1.full

多组学数据整合

bioRxiv CMOT: Cross Modality Optimal Transport for multimodal inference

Highlights: "It may not always be feasible to conduct such profiling experiments (e.g., missing modalities). Furthermore, even with the available multimodalities, data integration remains elusive since modalities may not always have paired samples, leaving partial to no correspondence information. To address those challenges, we developed Cross-Modality Optimal Transport (CMOT), a computational approach to infer missing modalities of single cells based on optimal transport (OT)."

Cross-Modality Optimal Transport (CMOT)

Link: https://www.biorxiv.org/content/10.1101/2022.10.15.512387v1.full

bioRxiv JAMIE: Joint Variational Autoencoders for Multi-Modal Imputation and Embedding

Highlights: "Recently, machine learning approaches have been developed to impute cell data but typically use fully matched multi-modal data and learn common latent embeddings that potentially miss modality specificity. To address these issues, we developed JAMIE takes single-cell multi-modal data that can have partially matched samples across modalities. VAE embeddings from matched samples across modalities are aggregated to identify joint cross-modal latent embeddings before reconstruction. To perform cross-modal imputation from one to another, the latent embeddings can be used with the opposite decoder."

JAMIE: Joint variational Autoencoders for Multi-modal Imputation and Embedding

Link: https://www.biorxiv.org/content/10.1101/2022.10.15.512388v1.full

单细胞

bioRxiv Deep generative modeling for quantifying sample-level heterogeneity in single-cell omics

Highlights: "Current approaches for analyzing condition-level heterogeneity in these experiments often rely on a simplification of the data such as an aggregation at the cell-type or cell-state-neighborhood level. Here we present MrVI, a deep generative model that provides sample-sample comparisons at a single-cell resolution, permitting the discovery of subtle sample-specific effects across cell populations. Additionally, the output of MrVI can be used to quantify the association between sample-level metadata and cell state variation."

Overview of MrVI.

Link: https://www.biorxiv.org/content/10.1101/2022.10.04.510898v1.full

空间转录组

Nature Biotechnology Modeling intercellular communication in tissues using spatial graphs of cells

Highlights: "Models of intercellular communication in tissues are based on molecular profiles of dissociated cells, are limited to receptor–ligand signaling and ignore spatial proximity in situ. We present node-centric expression modeling, a method based on graph neural networks that estimates the effects of niche composition on gene expression in an unbiased manner from spatial molecular profiling data. We recover signatures of molecular processes known to underlie cell communication."

Node-centric expression models capture statistical dependencies between cells in space.

Link: https://www.nature.com/articles/s41587-022-01467-z

Genome Biology Evaluation of cell-cell interaction methods by integrating single-cell RNA sequencing data with spatial information

Highlights: "We benchmark 16 cell-cell interaction methods by integrating scRNA-seq with ST data. We characterize cell-cell interactions into short-range and long-range interactions using spatial distance distributions between ligands and receptors. Our results suggest that the interactions predicted by different tools are highly dynamic, and the statistical-based methods show overall better performance than network-based methods and ST-based methods."

Defining and validating short-range and long-range interactions.

Link: https://genomebiology.biomedcentral.com/articles/10.1186/s13059-022-02783-y

bioRxiv St2cell: Reconstruction of in situ single-cell spatial transcriptomics by integrating high-resolution histological image

Highlights: "Here, to in silico reconstruct Spatially resolved transcriptomics (SRT) at the single-cell resolution, we propose St2cell which combines deep learning-based frameworks with a novel convex quadratic programming (CQP)-based model. St2cell can thoroughly leverage information in high-resolution (HR) histological images, enabling the accurate segmentation of in situ single cells and identification of their transcriptomics."

Overview of the St2cell.

Link: https://www.biorxiv.org/content/10.1101/2022.10.13.512059v1.full

序列分析

bioRxiv GSearch: Ultra-Fast and Scalable Microbial Genome Search by combining Kmer Hashing with Hierarchical Navigable Small World Graphs

Highlights: "Genome search and/or classification is a key step in microbiome studies and has become more challenging due to the increasing number of available (reference) genomes. By combining a kmer hashing-based genomic distance metric (Probminhash) with a graph based nearest neighbor search (NNS) algorithm (called Hierarchical Navigable Small World Graphs), we developed GSearch, that is at least ten times faster than alternative tools for the same purposes while maintaining high accuracy."

Schematic overview of GSearch building graph and searching graph steps.

Link: https://www.biorxiv.org/content/10.1101/2022.10.21.513218v1.full

bioRxiv streammd: fast low-memory duplicate marking using a Bloom filter

Highlights: "For large sequencing libraries, identification of duplicate reads is typically time-consuming and resource-intensive. Here we present streammd: a fast, memory-efficient, single-pass duplicate marking tool operating on the principle of a Bloom filter. We show that streammd closely reproduces the outputs of Picard MarkDuplicates, while being substantially faster and suitable for pipelined applications, and that it requires much less memory than SAMBLASTER, another single-pass duplicate marking tool."

Link: https://www.biorxiv.org/content/10.1101/2022.10.12.511997v1.full

其它关注

bioRxiv MorphNet Predicts Cell Morphology from Single-Cell Gene Expression

Highlights: "We present MorphNet, a computational approach that can draw pictures of a cell’s morphology from its gene expression profile. We employ state-of-the-art data augmentation techniques that allow training using as few as 103 images."

Schematic of MorphNet for predicting cell morphology from gene expression.

Link: https://www.biorxiv.org/content/10.1101/2022.10.21.513201v1.full

Nature Nonlinear decision-making with enzymatic neural networks

Highlights: "Artificial neural networks have revolutionized electronic computing. Similarly, molecular networks with neuromorphic architectures may enable molecular decision-making on a level comparable to gene regulatory networks. Here we introduce DNA-encoded enzymatic neurons with tuneable weights and biases, and which are assembled in multilayer architectures to classify nonlinearly separable regions."

Architecture of DNA-encoded enzymatic neural networks.

Link: https://www.nature.com/articles/s41586-022-05218-7

《生信了》2022年11月