SingleCellGGM enables gene expression program identification from single-cell transcriptomes and facilitates universal cell label transfer.

Gene co-expression analysis of single-cell transcriptomes, aiming to define functional relationships between genes, is challenging due to excessive dropout values. Here, we developed a single-cell graphical Gaussian model (SingleCellGGM) algorithm to conduct single-cell gene co-expression network analysis. When applied to mouse single-cell datasets, SingleCellGGM constructed networks from which gene co-expression modules with highly significant functional enrichment were identified. We considered the modules as gene expression programs (GEPs). These GEPs enable direct cell-type annotation of individual cells without cell clustering, and they are enriched with genes required for the functions of the corresponding cells, sometimes at levels greater than 10-fold. The GEPs are conserved across datasets and enable universal cell-type label transfer across different studies. We also proposed a dimension-reduction method through averaging by GEPs for single-cell analysis, enhancing the interpretability of results. Thus, SingleCellGGM offers a unique GEP-based perspective to analyze single-cell transcriptomes and reveals biological insights shared by different single-cell datasets.

Production of Aromatic Hydrocarbons from Co-Hydropyrolysis of Biomass Components and HDPE with Application of Modified HZSM-5 Catalyst.

Experiments were conducted in this study on the co-hydropyrolysis of three components of biomass (cellulose, hemicellulose, and lignin) and HDPE by using SR-Pd/Trap-HZ-5 as catalyst. To control the variable, we use the same experiment conditions in co-hydropyrolysis: Si/Al ratio of 50, Pd load 1 %, catalyst to reactant ratio of 1 : 10, 1 MPa, 400 °C, reaction time 1 h. Use XRD, TEM, BET, and NH3-TPD to confirm catalyst successful synthesis; use pine sawdust (PW) co-hydropyrolysis with HDPE to analyse catalytic activity; and use GC/MS to characterize the chemical composition of the bio-oil from the co-hydropyrolysis of biomass components and HDPE. The results show that cellulose has a significant synergistic effect with aromatic hydrocarbon production, whose selectivity was 93.3 %; hemicellulose has a synergistic effect; aromatic selectivity can reach 75.1 %; and a negative synergistic effect between lignin and HDPE was shown as the selectivity of aromatic hydrocarbons decreased from 62.1 % to 15.6 %.

Rice co-expression network analysis identifies gene modules associated with agronomic traits.

Identifying trait-associated genes is critical for rice (Oryza sativa) improvement, which usually relies on map-based cloning, quantitative trait locus analysis, or genome-wide association studies. Here we show that trait-associated genes tend to form modules within rice gene co-expression networks, a feature that can be exploited to discover additional trait-associated genes using reverse genetics. We constructed a rice gene co-expression network based on the graphical Gaussian model using 8,456 RNA-seq transcriptomes, which assembled into 1,286 gene co-expression modules functioning in diverse pathways. A number of the modules were enriched with genes associated with agronomic traits, such as grain size, grain number, tiller number, grain quality, leaf angle, stem strength, and anthocyanin content, and these modules are considered to be trait-associated gene modules. These trait-associated gene modules can be used to dissect the genetic basis of rice agronomic traits and to facilitate the identification of trait genes. As an example, we identified a candidate gene, OCTOPUS-LIKE 1 (OsOPL1), a homolog of the Arabidopsis (Arabidopsis thaliana) OCTOPUS gene, from a grain size module and verified it as a regulator of grain size via functional studies. Thus, our network represents a valuable resource for studying trait-associated genes in rice.

An Arabidopsis expression predictor enables inference of transcriptional regulators for gene modules.

The regulation of gene expression by transcription factors (TFs) has been studied for a long time, but no model that can accurately predict transcriptome profiles based on TF activities currently exists. Here, we developed a computational approach, named EXPLICIT (Expression Prediction via Log-linear Combination of Transcription Factors), to construct a universal predictor for Arabidopsis to predict the expression of 29 182 non-TF genes using 1678 TFs. When applied to RNA-Seq samples from diverse tissues, EXPLICIT generated accurate predicted transcriptomes correlating well with actual expression, with an average correlation coefficient of 0.986. After recapitulating the quantitative relationships between TFs and their target genes, EXPLICIT enabled downstream inference of TF regulators for genes and gene modules functioning in diverse plant pathways, including those involved in suberin, flavonoid, glucosinolate metabolism, lateral root, xylem, secondary cell wall development or endoplasmic reticulum stress response. Our approach showed a better ability to recover the correct TF regulators when compared with existing plant tools, and provides an innovative way to study transcriptional regulation.