Identifying signaling genes in spatial single cell expression data

Motivation Recent technological advances enable the profiling of spatial single cell expression data. Such data presents a unique opportunity to study cell-cell interactions and the signaling genes that mediate them. However, most current methods for the analysis of this data focus on unsupervised descriptive modeling, making it hard to identify key signaling genes and quantitatively assess their impact. Results We developed a Mixture of Experts for Spatial Signaling genes Identification (MESSI) method to identify active signaling genes within and between cells. The mixture of experts strategy enables MESSI to subdivide cells into subtypes. MESSI relies on multi-task learning using information from neighboring cells to improve the prediction of response genes within a cell. Applying the methods to three spatial single cell expression datasets, we show that MESSI accurately predicts the levels of response genes, improving upon prior methods and provides useful biological insights about key signaling genes and subtypes of excitatory neuron cells. Availability MESSI is available at: https://github.com/doraadong/MESSI Contact zivbj@cs.cmu.edu

[1]  B. Mattson,et al.  Preference for cocaine- versus pup-associated cues differentially activates neurons expressing either Fos or cocaine- and amphetamine-regulated transcript in lactating, maternal rodents , 2005, Neuroscience.

[2]  Shiquan Sun,et al.  Statistical analysis of spatial expression patterns for spatially resolved transcriptomic studies , 2020, Nature Methods.

[3]  Michael I. Jordan,et al.  Hierarchical Mixtures of Experts and the EM Algorithm , 1994, Neural Computation.

[4]  Mark S Cembrowski,et al.  Single-cell transcriptomics as a framework and roadmap for understanding the brain , 2019, Journal of Neuroscience Methods.

[5]  Koki Tsuyuzaki,et al.  Uncovering hypergraphs of cell-cell interaction from single cell RNA-sequencing data , 2019, bioRxiv.

[6]  D. Pfaff,et al.  Handbook of neuroendocrinology , 2012 .

[7]  Jacques Colinge,et al.  SingleCellSignalR: inference of intercellular networks from single-cell transcriptomics , 2020, Nucleic acids research.

[8]  Lars E. Borm,et al.  The promise of spatial transcriptomics for neuroscience in the era of molecular cell typing , 2017, Science.

[9]  William E. Allen,et al.  Three-dimensional intact-tissue sequencing of single-cell transcriptional states , 2018, Science.

[10]  J. Ajani,et al.  iTALK: an R Package to Characterize and Illustrate Intercellular Communication , 2019, bioRxiv.

[11]  Chenglong Xia,et al.  Spatial transcriptome profiling by MERFISH reveals subcellular RNA compartmentalization and cell cycle-dependent gene expression , 2019, Proceedings of the National Academy of Sciences.

[12]  Nimrod D. Rubinstein,et al.  Molecular, spatial, and functional single-cell profiling of the hypothalamic preoptic region , 2018, Science.

[13]  Theo Knijnenburg,et al.  Extracting Intercellular Signaling Network of Cancer Tissues using Ligand-Receptor Expression Patterns from Whole-tumor and Single-cell Transcriptomes , 2017, Scientific Reports.

[14]  Hal Daumé,et al.  Simultaneously Leveraging Output and Task Structures for Multiple-Output Regression , 2012, NIPS.

[15]  Rickard Sandberg,et al.  Identification of spatial expression trends in single-cell gene expression data , 2018, Nature Methods.

[16]  Piero Carninci,et al.  A draft network of ligand–receptor-mediated multicellular signalling in human , 2015, Nature Communications.

[17]  S. Yen,et al.  Reproductive Endocrinology: Physiology, Pathophysiology, and Clinical Management , 1999 .

[18]  W. Armstrong,et al.  Oxytocin release in magnocellular nuclei: neurochemical mediators and functional significance during gestation. , 2010, American journal of physiology. Regulatory, integrative and comparative physiology.

[19]  A. Regev,et al.  Scaling single-cell genomics from phenomenology to mechanism , 2017, Nature.

[20]  Guocheng Yuan,et al.  Identification of spatially associated subpopulations by combining scRNA-seq and sequential fluorescence in situ hybridization data , 2018, Nature Biotechnology.

[21]  O. Bosch,et al.  More than reproduction: Central gonadotropin‐releasing hormone antagonism decreases maternal aggression in lactating rats , 2019, Journal of neuroendocrinology.

[22]  Douglas A. Lauffenburger,et al.  Analysis of Single-Cell RNA-Seq Identifies Cell-Cell Communication Associated with Tumor Characteristics , 2018, Cell reports.

[23]  S. Lightman,et al.  Synchronous activation of gonadotropin-releasing hormone gene transcription and secretion by pulsatile kisspeptin stimulation , 2013, Proceedings of the National Academy of Sciences.

[24]  Kan Liu,et al.  Giotto, a toolbox for integrative analysis and visualization of spatial expression data , 2020 .

[25]  Tianqi Chen,et al.  XGBoost: A Scalable Tree Boosting System , 2016, KDD.

[26]  Long Cai,et al.  Giotto, a pipeline for integrative analysis and visualization of single-cell spatial transcriptomic data , 2019, bioRxiv.

[27]  H. Kanasaki,et al.  Pulse Frequency-Dependent Gonadotropin Gene Expression by Adenylate Cyclase-Activating Polypeptide 1 in Perifused Mouse Pituitary Gonadotroph LbetaT2 Cells1 , 2009, Biology of reproduction.

[28]  D. Pow,et al.  Dendrites of hypothalamic magnocellular neurons release neurohypophysial peptides by exocytosis , 1989, Neuroscience.

[29]  Mirjana Efremova,et al.  CellPhoneDB: inferring cell–cell communication from combined expression of multi-subunit ligand–receptor complexes , 2020, Nature Protocols.

[30]  S. Teichmann,et al.  SpatialDE: identification of spatially variable genes , 2018, Nature Methods.

[31]  Han Zhang,et al.  Gene Expression Value Prediction Based on XGBoost Algorithm , 2019, Front. Genet..

[32]  James K. Rilling,et al.  The biology of mammalian parenting and its effect on offspring social development , 2014, Science.

[33]  Yi Li,et al.  Gene expression inference with deep learning , 2015, bioRxiv.