Underlying Gene Regulatory Network Research Articles

Stochastic modeling of a gene regulatory network driving B cell development in germinal centers.

Germinal centers (GCs) are the key histological structures of the adaptive immune system, responsible for the development and selection of B cells producing high-affinity antibodies against antigens. Due to their level of complexity, unexpected malfunctioning may lead to a range of pathologies, including various malignant formations. One promising way to improve the understanding of malignant transformation is to study the underlying gene regulatory networks (GRNs) associated with cell development and differentiation. Evaluation and inference of the GRN structure from gene expression data is a challenging task in systems biology: recent achievements in single-cell (SC) transcriptomics allow the generation of SC gene expression data, which can be used to sharpen the knowledge on GRN structure. In order to understand whether a particular network of three key gene regulators (BCL6, IRF4, BLIMP1), influenced by two external stimuli signals (surface receptors BCR and CD40), is able to describe GC B cell differentiation, we used a stochastic model to fit SC transcriptomic data from a human lymphoid organ dataset. The model is defined mathematically as a piecewise-deterministic Markov process. We showed that after parameter tuning, the model qualitatively recapitulates mRNA distributions corresponding to GC and plasmablast stages of B cell differentiation. Thus, the model can assist in validating the GRN structure and, in the future, could lead to better understanding of the different types of dysfunction of the regulatory mechanisms.

Abstract 5853: Regulatory network inference of lung adenocarcinoma drug-tolerant persister cells

Abstract Despite the effectiveness of the latest generation of cancer drugs, partial tumor response and subsequent disease relapse is commonly observed in patients. Recent evidence suggests that treatment resistance emerges from drug-tolerant persister cells (DTPC) that survive through non-genetic tumor cell adaptation. Although DTPC are mostly slow-cycling, a rare subset can re-enter the cell cycle under constitutive presence of treatment. Understanding how DTPC dynamically remodel their molecular portraits at the single-cell level may provide high-resolution insights into the molecular circuits associated with adaptive survival to therapy.Here, we use single-cell RNA sequencing (scRNA-seq) to identify potential mechanisms that promote survival of DTPC and decipher their underlying gene regulatory networks (GRNs), in a model of EGFR-mutant lung adenocarcinoma treated with osimertinib (EGFR inhibitor). Gene expression analysis revealed fewer cell states in DTP compared to treatment-naïve cells (TN). We confirmed the acquisition of DTPC hallmarks (e.g., high epithelial-mesenchymal transition, slow-cycling phenotype, deficient DNA repair). We also identified subpopulations of cells that display intermediate transcriptional states: a pre-existing DTP-like state in a bulk of TN cells, and DTPC with enriched cell cycle signatures, evoking a phenotype of cycling DTPC. To characterize the GRN that govern TN cells, DTPC and the intermediate cell states, we applied the network inference tool PYSCENIC, which associates transcription factors (TFs) to their candidate target genes (TGs) on the basis of the scRNA-seq data and enrichment of TF binding motifs in the genomic regions around TGs. The resulting (GRN) consisted of 5 TF modules clustered on the basis of correlated TF activity.This analysis revealed that, compared to TN cells, DTPC and the intermediate cell states display a marked decreased expression of TFs and TGs involved in mRNA splicing, and more specifically in the initiation of spliceosome assembly through E-complex and A-complex formation. Interestingly, the splicing-related TFs all appear to have affinity for binding to a GCCxC consensus motif or close variant in genes (i.e., SNRPB and SNRPE) that code for key effectors of these initiation complexes. This pilot study highlights that scRNA-seq analyses coupled with inference of regulatory networks might enable the identification of molecular mechanisms that potentially trigger cell state transitions during drug-tolerance. The role of altered mRNA splicing in persistence to EGFR-targeted therapies will be explored experimentally in patient-derived cell lines and 3D organoid cultures. Citation Format: Lisa Chabrier, Camille Leonce, Raphael Schneider, Cyril Degletagne, Pierre Saintigny, Anton Crombach, Sandra Ortiz-Cuaran. Regulatory network inference of lung adenocarcinoma drug-tolerant persister cells [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 5853.

Reconstructing growth and dynamic trajectories from single-cell transcriptomics data

Time-series single-cell RNA sequencing (scRNA-seq) datasets provide unprecedented opportunities to learn dynamic processes of cellular systems. Due to the destructive nature of sequencing, it remains challenging to link the scRNA-seq snapshots sampled at different time points. Here we present TIGON, a dynamic, unbalanced optimal transport algorithm that reconstructs dynamic trajectories and population growth simultaneously as well as the underlying gene regulatory network from multiple snapshots. To tackle the high-dimensional optimal transport problem, we introduce a deep learning method using a dimensionless formulation based on the Wasserstein–Fisher–Rao (WFR) distance. TIGON is evaluated on simulated data and compared with existing methods for its robustness and accuracy in predicting cell state transition and cell population growth. Using three scRNA-seq datasets, we show the importance of growth in the temporal inference, TIGON’s capability in reconstructing gene expression at unmeasured time points and its applications to temporal gene regulatory networks and cell–cell communication inference.

Quantifying cancer cell plasticity with gene regulatory networks and single-cell dynamics.

Phenotypic plasticity of cancer cells can lead to complex cell state dynamics during tumor progression and acquired resistance. Highly plastic stem-like states may be inherently drug-resistant. Moreover, cell state dynamics in response to therapy allow a tumor to evade treatment. In both scenarios, quantifying plasticity is essential for identifying high-plasticity states or elucidating transition paths between states. Currently, methods to quantify plasticity tend to focus on 1) quantification of quasi-potential based on the underlying gene regulatory network dynamics of the system; or 2) inference of cell potency based on trajectory inference or lineage tracing in single-cell dynamics. Here, we explore both of these approaches and associated computational tools. We then discuss implications of each approach to plasticity metrics, and relevance to cancer treatment strategies.

Opposing gene regulatory programs governing myofiber development and maturation revealed at single nucleus resolution

Skeletal muscle fibers express distinct gene programs during development and maturation, but the underlying gene regulatory networks that confer stage-specific myofiber properties remain unknown. To decipher these distinctive gene programs and how they respond to neural activity, we generated a combined multi-omic single-nucleus RNA-seq and ATAC-seq atlas of mouse skeletal muscle development at multiple stages of embryonic, fetal, and postnatal life. We found that Myogenin, Klf5, and Tead4 form a transcriptional complex that synergistically activates the expression of muscle genes in developing myofibers. During myofiber maturation, the transcription factor Maf acts as a transcriptional switch to activate the mature fast muscle gene program. In skeletal muscles of mutant mice lacking voltage-gated L-type Ca2+ channels (Cav1.1), Maf expression and myofiber maturation are impaired. These findings provide a transcriptional atlas of muscle development and reveal genetic links between myofiber formation, maturation, and contraction.

Correlated stabilizing selection shapes the topology of gene regulatory networks.

The evolution of gene expression is constrained by the topology of gene regulatory networks, as co-expressed genes are likely to have their expressions affected together by mutations. Conversely, co-expression can also be an advantage when genes are under joint selection. Here, we assessed theoretically whether correlated selection (selection for a combination of traits) was able to affect the pattern of correlated gene expressions and the underlying gene regulatory networks. We ran individual-based simulations, applying a stabilizing correlated fitness function to three genetic architectures: a quantitative genetics (multilinear) model featuring epistasis and pleiotropy, a quantitative genetics model where each genes has an independent mutational structure, and a gene regulatory network model, mimicking the mechanisms of gene expression regulation. Simulations showed that correlated mutational effects evolved in the three genetic architectures as a response to correlated selection, but the response in gene networks was specific. The intensity of gene co-expression was mostly explained by the regulatory distance between genes (largest correlations being associated to genes directly interacting with each other), and the sign of co-expression was associated with the nature of the regulation (transcription activation or inhibition). These results concur to the idea that gene network topologies could partly reflect past selection patterns on gene expression.

Abstract 5371: Modeling single-cell dynamics using stochastic generative models based on neural differential equations

Abstract Hematopoietic malignancies arise from driver alterations acquired during specific, often intermediate and transient, differentiation states. When adequately sampled, “snapshot” single-cell measurements can capture dynamic biological processes, including transient states. While there are effective computational modeling solutions to order and capture the relationships among these cell states (often projecting onto a lower dimensional manifold with directionality), methods to infer the causal gene regulatory mechanisms driving cell state transitions remain limited. Recently, a mathematical framework for modeling cell dynamics from snapshot data using a physics-based drift-diffusion equation was proposed. This drift-diffusion equation framework describes the dynamics of cell distributions with respect to time, wherein stereotypical driving forces (e.g., hematopoietic development) are captured by the drift term, and biological stochasticity is attributed to the diffusion term. The existing modeling solutions within this framework are forced to assume a fixed diffusion parameter across cell states, ascribing all observed dynamics to drift alone. Here, we expand on current drift-diffusion models to learn both the drift and diffusion terms that describe every observed cell state. These models are constructed from stochastic neural differential equations, continuous deep neural networks that approximate the theoretically complex differential equations underlying gene regulatory networks in hematopoiesis. Lineage-barcoded, multi-time point single-cell RNA sequencing experiments can approximate ground truth cellular dynamics. Using such data, we demonstrate the predictive accuracy of our proposed approach through benchmarking against the state-of-the-art methods. Through our drift-diffusion model, we can now distinguish, for the first time, between the deterministic and stochastic contributions to cell fate decision making. In doing so, we identify genes whose expression is associated with cellular drift and indeed show that they correspond to well-known markers of specific cell types; moreover, we identify genes that are associated with cell states that exhibit notable diffusion. Finally, we demonstrate the potential of such models to capture the essence of the dynamic processes through prediction on out-of-distribution data (referred to as “transfer learning” in the machine learning field). For example, when the model is trained on in vitro hematopoiesis data, our model is able to accurately predict cell trajectories in the corresponding in vivo mouse model. We anticipate that such insights will be useful in identifying key dynamic and time-dependent regulatory processes in normal development and cancer. Citation Format: Michael E. Vinyard, Anders W. Rasmussen, Ruitong Li, Luca Pinello, Gad Getz. Modeling single-cell dynamics using stochastic generative models based on neural differential equations. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 5371.

Characterization of eyes, photoreceptors, and opsins in developmental stages of the arrow worm Spadella cephaloptera (Chaetognatha).

The phylogenetic position of chaetognaths, or arrow worms, has been debated for decades, however recently they have been grouped into the Gnathifera, a sister clade to all other Spiralia. Chaetognath photoreceptor cells are anatomically unique by exhibiting a highly modified cilium and are arranged differently in the eyes of the various species. Studies investigating eye development and underlying gene regulatory networks are so far missing. To gain insights into the development and the molecular toolkit of chaetognath photoreceptors and eyes a new transcriptome of the epibenthic species Spadella cephaloptera was searched for opsins. Our screen revealed two copies of xenopsin and a single copy of peropsin. Gene expression analyses demonstrated that only xenopsin1 is expressed in photoreceptor cells of the developing lateral eyes. Adults likewise exhibit two xenopsin1 + photoreceptor cells in each of their lateral eyes. Beyond that, a single cryptochrome gene was uncovered and found to be expressed in photoreceptor cells of the lateral developing eye. In addition, cryptochrome is also expressed in the cerebral ganglia in a region in which also peropsin expression was observed. This condition is reminiscent of a nonvisual photoreceptive zone in the apical nervous system of the annelid Platynereis dumerilii that performs circadian entrainment and melatonin release. Cryptochrome is also expressed in cells of the corona ciliata, an organ in the posterior dorsal head region, indicating a role in circadian entrainment. Our study highlights the importance of the Gnathifera for unraveling the evolution of photoreceptors and eyes in Spiralia and Bilateria.

Single-cell and spatial transcriptomics reveal aberrant lymphoid developmental programs driving granuloma formation

Granulomas are lumps of immune cells that can form in various organs. Most granulomas appear unstructured, yet they have some resemblance to lymphoid organs. To better understand granuloma formation, we performed single-cell sequencing and spatial transcriptomics on granulomas from patients with sarcoidosis and bioinformatically reconstructed the underlying gene regulatory networks. We discovered an immune stimulatory environment in granulomas that repurposes transcriptional programs associated with lymphoid organ development. Granuloma formation followed characteristic spatial patterns and involved genes linked to immunometabolism, cytokine and chemokine signaling, and extracellular matrix remodeling. Three cell types emerged as key players in granuloma formation: metabolically reprogrammed macrophages, cytokine-producing Th17.1 cells, and fibroblasts with inflammatory and tissue-remodeling phenotypes. Pharmacological inhibition of one of the identified processes attenuated granuloma formation in a sarcoidosis mouse model. We show that human granulomas adopt characteristic aspects of normal lymphoid organ development in aberrant combinations, indicating that granulomas constitute aberrant lymphoid organs.

Comparative expression analysis in three Brassicaceae species revealed compensatory changes of the underlying gene regulatory network.

Trichomes are regularly distributed on the leaves of Arabidopsis thaliana. The gene regulatory network underlying trichome patterning involves more than 15 genes. However, it is possible to explain patterning with only five components. This raises the questions about the function of the additional components and the identification of the core network. In this study, we compare the relative expression of all patterning genes in A. thaliana, A. alpina and C. hirsuta by qPCR analysis and use mathematical modelling to determine the relative importance of patterning genes. As the involved proteins exhibit evolutionary conserved differential complex formation, we reasoned that the genes belonging to the core network should exhibit similar expression ratios in different species. However, we find several striking differences of the relative expression levels. Our analysis of how the network can cope with such differences revealed relevant parameters that we use to predict the relevant molecular adaptations in the three species.

Improving crop yield potential: Underlying biological processes and future prospects.

The growing world population and global increases in the standard of living both result in an increasing demand for food, feed and other plant-derived products. In the coming years, plant-based research will be among the major drivers ensuring food security and the expansion of the bio-based economy. Crop productivity is determined by several factors, including the available physical and agricultural resources, crop management, and the resource use efficiency, quality and intrinsic yield potential of the chosen crop. This review focuses on intrinsic yield potential, since understanding its determinants and their biological basis will allow to maximize the plant's potential in food and energy production. Yield potential is determined by a variety of complex traits that integrate strictly regulated processes and their underlying gene regulatory networks. Due to this inherent complexity, numerous potential targets have been identified that could be exploited to increase crop yield. These encompass diverse metabolic and physical processes at the cellular, organ and canopy level. We present an overview of some of the distinct biological processes considered to be crucial for yield determination that could further be exploited to improve future crop productivity.

ΔNp63/p73 drive metastatic colonization by controlling a regenerative epithelial stem cell program in quasi-mesenchymal cancer stem cells.

Cancer stem cells (CSCs) may serve as the cellular seeds of tumor recurrence and metastasis, and they can be generated via epithelial-mesenchymal transitions (EMTs). Isolating pure populations of CSCs is difficult because EMT programs generate multiple alternative cell states, and phenotypic plasticity permits frequent interconversions between these states. Here, we used cell-surface expression of integrin β4 (ITGB4) to isolate highly enriched populations of human breast CSCs, and we identified the gene regulatory network operating in ITGB4+ CSCs. Specifically, we identified ΔNp63 and p73, the latter of which transactivates ΔNp63, as centrally important transcriptional regulators of quasi-mesenchymal CSCs that reside in an intermediate EMT state. We found that the transcriptional program controlled by ΔNp63 in CSCs is largely distinct from the one that it orchestrates in normal basal mammary stem cells and, instead, it more closely resembles a regenerative epithelial stem cell response to wounding. Moreover, quasi-mesenchymal CSCs repurpose this program to drive metastatic colonization via autocrine EGFR signaling.

CNSC-33. TRANSCRIPTOMIC NEURON-TUMOR INTERACTION MAPPING OF EXTRACRANIAL TUMORS AND BRAIN METASTASES

Abstract Ongoing efforts in the field of Cancer Neuroscience seek to improve the understanding of neuron-cancer interactions and their effects on tumor progression. In glioma, synaptic neuronal input to tumor cells drives tumor progression and invasion. Even though neuron-cancer interactions have been described in brain metastases and extracranial tumors as well, the mechanisms of action between the nervous system and tumor cells are incompletely understood. Systematic transcriptomic analyses to detect overarching principles of neuron-tumor interaction have been lacking so far. Here, we performed focused integrative transcriptomic and proteomic analyses in more than 10 different extracranial tumor entities and brain metastases to analyse potential neuron-tumor interactions.Neurotransmitters receptor genes were heterogeneously expressed across extracranial tumor entities and brain metastases. We used Markov Affinity-based Graph Imputation of Cells to recover gene-gene relationships to robustly identify more than 50 initial gene co-expression networks (GCNs) associated with neurotransmitter genes. Molecular pathways that are associated with neurotransmitter receptor gene expression show heterogenous and overlapping neuronal gene expression programs in tumor cell subpopulations. While some GCNs showed overarching expression patterns of neuronal interactions across many entities including brain metastases as well as extracranial tumors, some GCNs were specialised signatures for certain tumor types. To characterise underlying gene regulatory networks, we performed transcription factor analyses. Certain GCNs were clearly associated with specific transcriptional regulators and known tumor biological functions. To validate our results, we further analysed the GCNs in bulk RNA datasets as well as on a protein level. We were able to identify neuronal gene expression patterns that were associated with worse survival across certain tumor entities.The results suggest that brain metastases and extracranial tumors have multiple biologically relevant mechanisms of interaction with the nervous system. These analyses are the foundation for further validation and clinical translation.

Decoding the IGF1 signaling gene regulatory network behind alveologenesis from a mouse model of bronchopulmonary dysplasia.

Lung development is precisely controlled by underlying gene regulatory networks (GRN). Disruption of genes in the network can interrupt normal development and cause diseases such as bronchopulmonary dysplasia (BPD) - a chronic lung disease in preterm infants with morbid and sometimes lethal consequences characterized by lung immaturity and reduced alveolarization. Here, we generated a transgenic mouse exhibiting a moderate severity BPD phenotype by blocking IGF1 signaling in secondary crest myofibroblasts (SCMF) at the onset of alveologenesis. Using approaches mirroring the construction of the model GRN in sea urchin's development, we constructed the IGF1 signaling network underlying alveologenesis using this mouse model that phenocopies BPD. The constructed GRN, consisting of 43 genes, provides a bird's eye view of how the genes downstream of IGF1 are regulatorily connected. The GRN also reveals a mechanistic interpretation of how the effects of IGF1 signaling are transduced within SCMF from its specification genes to its effector genes and then from SCMF to its neighboring alveolar epithelial cells with WNT5A and FGF10 signaling as the bridge. Consistently, blocking WNT5A signaling in mice phenocopies BPD as inferred by the network. A comparative study on human samples suggests that a GRN of similar components and wiring underlies human BPD. Our network view of alveologenesis is transforming our perspective to understand and treat BPD. This new perspective calls for the construction of the full signaling GRN underlying alveologenesis, upon which targeted therapies for this neonatal chronic lung disease can be viably developed.

Regulatory chromatin rewiring promotes metabolic switching during adaptation to oncogenic receptor tyrosine kinase inhibition

Oesophageal adenocarcinoma (OAC) patients show poor survival rates and there are few targeted molecular therapies available. However, components of the receptor tyrosine kinase (RTK) driven pathways are commonly mutated in OAC, typified by high frequency amplifications of the RTK ERBB2. ERBB2 can be therapeutically targeted, but this has limited clinical benefit due to the acquisition of drug resistance. Here we examined how OAC cells adapt to ERBB2 inhibition as they transition to a drug resistant state. ERBB2 inhibition triggers widespread remodelling of the accessible chromatin landscape and the underlying gene regulatory networks. The transcriptional regulators HNF4A and PPARGC1A play a key role in this network rewiring. Initially, inhibition of cell cycle associated gene expression programmes is observed, with compensatory increases in the programmes driving changes in metabolic activity. Both PPARGC1A and HNF4A are required for the acquisition of resistance to ERBB2 inhibition and PPARGC1A is instrumental in promoting a switch to dependency on oxidative phosphorylation. Our work therefore reveals the molecular pathways that support the acquisition of a resistant state and points to potential new therapeutic strategies to combat cellular adaptation and ensuing drug resistance.

Human stem cell models to study placode development, function and pathology.

Placodes are embryonic structures originating from the rostral ectoderm that give rise to highly diverse organs and tissues, comprising the anterior pituitary gland, paired sense organs and cranial sensory ganglia. Their development, including the underlying gene regulatory networks and signalling pathways, have been for the most part characterised in animal models. In this Review, we describe how placode development can be recapitulated by the differentiation of human pluripotent stem cells towards placode progenitors and their derivatives, highlighting the value of this highly scalable platform as an optimal in vitro tool to study the development of human placodes, and identify human-specific mechanisms in their development, function and pathology.

Phenotypic and transcriptomic responses of the shade-grown species Panax ginseng to variable light conditions.

Elucidating how plant species respond to variable light conditions is important to understand the ecological adaptation to heterogeneous natural habitats. Plant performance and its underlying gene regulatory network have been well documented in sun-grown plants. However, the phenotypic and molecular responses of shade-grown plants under variable light conditions have remained largely unclear. We assessed the differences in phenotypic performance between Panax ginseng (shade-grown) and Arabidopsis thaliana (sun-grown) under sunlight, shade and deep-shade conditions. To further address the molecular bases underpinning the phenotypic responses, we compared time-course transcriptomic expression profiling and candidate gene structures between the two species. Our results show that, compared with arabidopsis, ginseng plants not only possess a lower degree of phenotypic plasticity among the three light conditions, but also exhibit higher photosynthetic efficiency under shade and deep-shade conditions. Further comparisons of the gene expression and structure reveal that differential transcriptional regulation together with increased copy number of photosynthesis-related genes (e.g. electron transfer and carbon fixation) may improve the photosynthetic efficiency of ginseng plants under the two shade conditions. In contrast, the inactivation of phytochrome-interacting factors (i.e. absent and no upregulation of the PIF genes) are potentially associated with the observed low degree of phenotypic plasticity of ginseng plants under variable light conditions. Our study provides new insights into how shade-grown plants respond to variable light conditions. Candidate genes related to shade adaptation in ginseng provide valuable genetic resources for future molecular breeding of high-density planting crops.

Trichomes on female reproductive tract: rapid diversification and underlying gene regulatory network in Drosophila suzukii and its related species

BackgroundThe ovipositors of some insects are external female genitalia, which have their primary function to deliver eggs. Drosophila suzukii and its sibling species D. subpulchrella are known to have acquired highly sclerotized and enlarged ovipositors upon their shifts in oviposition sites from rotting to ripening fruits. Inside the ovipositor plates, there are scale-like polarized protrusions termed “oviprovector scales” that are likely to aid the mechanical movement of the eggs. The size and spatial distribution of the scales need to be rearranged following the divergence of the ovipositors. In this study, we examined the features of the oviprovector scales in D. suzukii and its closely related species. We also investigated whether the scales are single-cell protrusions comprised of F-actin under the same conserved gene regulatory network as the well-characterized trichomes on the larval cuticular surface.ResultsThe oviprovector scales of D. suzukii and D. subpulchrella were distinct in size and spatial arrangement compared to those of D. biarmipes and other closely related species. The scale numbers also varied greatly among these species. The comparisons of the size of the scales suggested a possibility that the apical cell area of the oviprovector has expanded upon the elongation of the ovipositor plates in these species. Our transcriptome analysis revealed that 43 out of the 46 genes known to be involved in the trichome gene regulatory network are expressed in the developing female genitalia of D. suzukii and D. subpulchrella. The presence of Shavenbaby (Svb) or svb was detected in the inner cavity of the developing ovipositors of D. melanogaster, D. suzukii, and D. subpulchrella. Also, shavenoid (sha) was expressed in the corresponding patterns in the developing ovipositors and showed differential expression levels between D. suzukii and D. subpulchrella at 48 h APF.ConclusionsThe oviprovector scales have divergent size and spatial arrangements among species. Therefore, these scales may represent a rapidly diversifying morphological trait of the female reproductive tract reflecting ecological contexts. Furthermore, our results showed that the gene regulatory network underlying trichome formation is also utilized to develop the rapidly evolving trichomes on the oviprovectors of these flies.

Data mining and mathematical models in cancer prognosis and prediction.

Cancer is a fetal and complex disease. Individual differences of the same cancer type or the same patient at different stages of cancer development may require distinct treatments. Pathological differences are reflected in tissues, cells and gene levels etc. The interactions between the cancer cells and nearby microenvironments can also influence the cancer progression and metastasis. It is a huge challenge to understand all of these mechanistically and quantitatively. Researchers applied pattern recognition algorithms such as machine learning or data mining to predict cancer types or classifications. With the rapidly growing and available computing powers, researchers begin to integrate huge data sets, multi-dimensional data types and information. The cells are controlled by the gene expressions determined by the promoter sequences and transcription regulators. For example, the changes in the gene expression through these underlying mechanisms can modify cell progressing in the cell-cycle. Such molecular activities can be governed by the gene regulations through the underlying gene regulatory networks, which are essential for cancer study when the information and gene regulations are clear and available. In this review, webriefly introduce several machine learning methods ofcancer prediction and classification which include Artificial Neural Networks (ANNs), Decision Trees (DTs), Support Vector Machine (SVM) and naive Bayes. Then we describe a few typical models for building up gene regulatory networks such as Correlation, Regression and Bayes methods based on available data. These methods can help on cancer diagnosis such as susceptibility, recurrence, survival etc. At last, we summarize and compare the modeling methods to analyze the development and progression of cancer through gene regulatory networks. These models can provide possible physical strategies to analyze cancer progression in a systematic and quantitative way.

Benchmarking imputation methods for network inference using a novel method of synthetic scRNA-seq data generation

BackgroundSingle cell RNA-sequencing (scRNA-seq) has very rapidly become the new workhorse of modern biology providing an unprecedented global view on cellular diversity and heterogeneity. In particular, the structure of gene-gene expression correlation contains information on the underlying gene regulatory networks. However, interpretation of scRNA-seq data is challenging due to specific experimental error and biases that are unique to this kind of data including drop-out (or technical zeros).MethodsTo deal with this problem several methods for imputation of zeros for scRNA-seq have been developed. However, it is not clear how these processing steps affect inference of genetic networks from single cell data. Here, we introduce Biomodelling.jl, a tool for generation of synthetic scRNA-seq data using multiscale modelling of stochastic gene regulatory networks in growing and dividing cells.ResultsOur tool produces realistic transcription data with a known ground truth network topology that can be used to benchmark different approaches for gene regulatory network inference. Using this tool we investigate the impact of different imputation methods on the performance of several network inference algorithms.ConclusionsBiomodelling.jl provides a versatile and useful tool for future development and benchmarking of network inference approaches using scRNA-seq data.