Cell systems最新文献

Microbiomes often show similar functional profiles despite highly variable taxonomic compositions, a phenomenon attributed to "functional redundancy" and presumed selection for functional traits. However, this link between functional variability and selection remains vaguely defined. We demonstrate that reduced functional variability can arise from statistical averaging when aggregating taxonomic abundances and does not necessarily imply selection. We introduce an empirical null model that accounts for this statistical averaging effect. Applying this model to microbial communities from bromeliad foliage, we find no evidence of functional selection. In contrast, soil and human gut communities grown in vitro exhibit selection for metabolic functions. We also find that correlations between functions and taxonomic abundances can produce misleading signals of selection. Using an extended null model, we show that apparent functional selection in Human Microbiome Project data is artifactual. Our framework clarifies the conditions under which functional selection can be meaningfully inferred from microbiome data.

Clostridioides difficile (C. difficile) colonizes up to 40% of community-dwelling adults without causing disease but can eventually lead to infection (C. difficile infection [CDI]). There has been a lack of focus on how to prevent colonization and facilitate the successful clearance of C. difficile prior to the emergence of CDI. We show that microbial community-scale metabolic models (MCMMs) accurately predict C. difficile colonization susceptibility in vitro and in vivo, offering mechanistic insights into microbiota-specific interactions involving metabolites like succinate, trehalose, and ornithine. MCMMs reveal distinct C. difficile metabolic niches-two growth-associated and one non-growth-associated-observed across 15,204 individuals from five cohorts. We further demonstrate that MCMMs can predict personalized C. difficile growth suppression by a probiotic cocktail designed to replace fecal microbiota transplants (FMTs) for the treatment of recurrent CDI, and we identify new probiotic targets for future validation. MCMMs represent a powerful framework for predicting pathogen colonization and assessing probiotic efficacy across diverse microbiota contexts. A record of this paper's transparent peer review process is included in the supplemental information.

Generative models can now design a diverse set of protein backbones, yet the quantification of distributional similarities of protein structure embeddings revealed that current models fail to capture the full spectrum of structural elements at different hierarchical levels. SHAPES (structural and hierarchical assessment of proteins with embedding similarity) quantifies these gaps and delivers a benchmark to guide next-generation protein design.

Treatments for tuberculosis remain lengthy, motivating a search for new drugs with novel mechanisms of action. However, it remains challenging to determine the direct targets of a drug and which disrupted cellular processes lead to bacterial killing. We developed a computational tool, DECIPHAER (decoding cross-modal information of pharmacologies via autoencoders), to select the important correlated transcriptional and morphological responses of Mycobacterium tuberculosis to treatment. By finding a reduced feature space, DECIPHAER highlighted essential features of cellular damage. DECIPHAER provides cell-death-relevant insight into uni-modal datasets, enabling interrogation of drug treatment responses for which transcriptional data are unavailable. Using morphological data alone with DECIPHAER, we discovered that respiration inhibition by the polypharmacological drugs SQ109 and BM212 can influence cell death more than their effects on the cell wall. This study demonstrates that DECIPHAER can extract the critical shared information from multi-modal measurements to identify cell-death-relevant mechanisms of TB drugs. A record of this paper's transparent peer review process is included in the supplemental information.

Short tandem repeats (STRs) are enriched in regulatory regions and can bind transcription factors (TFs), as shown for selected examples in vitro. Here, we use a library-based assay to systematically measure TF binding to STRs of 2-5 bp units within budding yeast cells. We examined STR binding by four TFs, including Msn2, and further tested six Msn2 mutants, including two that contained only the DNA-binding domain (DBD) or only the 642-aa intrinsically disordered region (IDR). We find substantial STR effects on motif-dependent and motif-independent binding, which varied between TFs. For Msn2, STR association was explained by the DBD binding at motif half-sites and the IDR favoring homopurine-homopyrimidine and AT-rich repeats. TF-preferred STRs are enriched in the human genome but remain at low frequency at yeast promoters. We discuss the implications of our results for understanding the role of STRs and their crosstalk with TF IDRs in regulating TF binding across genomes.

Exon skipping (ES) is the most prevalent form of alternative splicing and a hallmark of tumorigenesis, yet its functional roles remain underexplored. Here, we present a CRISPR-RfxCas13d-based platform for transcript-specific silencing of ES-derived isoforms using guide RNAs (gRNAs) targeting exon-exon junctions. We designed a transcriptome-wide gRNA library against 3,744 human ES events and conducted loss-of-function screens in colorectal cancer (CRC) cells in vitro and in vivo. This screen uncovered multiple ES events essential for CRC growth, notably HMGN3 Δ6, an isoform arising from exon 6 skipping, which enhanced tumor proliferation. Functional validation confirmed the oncogenic role of HMGN3 Δ6 and its necessity for CRC progression. Our study establishes CRISPR-RfxCas13d as a powerful tool for isoform-specific functional genomics and reveals a widespread, previously uncharacterized layer of tumor biology driven by ES. These findings position ES-derived transcripts as promising targets for therapeutic intervention in cancer.

Macrophages are innate immune cells involved in host defense. Dissecting the regulatory landscape that enables their swift and specific response to pathogens, we performed time-series analysis of gene expression and chromatin accessibility in murine macrophages exposed to various immune stimuli, and we functionally evaluated gene knockouts at scale using a combined CROP-seq and CITE-seq assay. We identified new roles of transcription regulators such as Spi1/PU.1 and JAK-STAT pathway members in immune cell homeostasis and response to pathogens. Macrophage activity was modulated by splicing proteins SFPQ and SF3B1, histone acetyltransferase EP300, cohesin subunit SMC1A, and mediator complex proteins MED8 and MED14. We further observed crosstalk among immune signaling pathways and identified molecular drivers of pathogen-induced dynamics. In summary, this study establishes a time-resolved regulatory map of pathogen response in macrophages, and it describes a broadly applicable method for dissecting immune-regulatory programs through integrative time-series analysis and high-content CRISPR screening. A record of this paper's transparent peer review process is included in the supplemental information.

Microbial colony growth is shaped by the physics of biomass propagation and nutrient diffusion and by the metabolic reactions that organisms activate as a function of the surrounding environment. While microbial colonies have been explored using minimal models of growth and motility, full integration of biomass propagation and metabolism is still lacking. Here, building upon our framework for computation of microbial ecosystems in time and space (COMETS), we combine dynamic flux balance modeling of metabolism with collective biomass propagation and demographic fluctuations to provide nuanced simulations of E. coli colonies. Simulations produced realistic colony morphology, consistent with our experiments. They characterize the transition between smooth and furcated colonies and the decay of genetic diversity. Furthermore, we demonstrate that under certain conditions, biomass can accumulate along "metabolic rings" that are reminiscent of coffee-stain rings but have a completely different origin. Our approach is a key step toward predictive microbial ecosystems modeling. A record of this paper's transparent peer review process is included in the supplemental information.

An important and largely unsolved problem in synthetic biology is how to target gene expression to specific cell types. Here, we apply iterative deep learning to design synthetic enhancers with strong differential activity between two human cell lines. We initially train models on published datasets of enhancer activity and chromatin accessibility and use them to guide the design of synthetic enhancers that maximize predicted specificity. We experimentally validate these sequences, use the measurements to re-optimize the model, and design a second generation of enhancers with improved specificity. Our design methods embed relevant transcription factor binding site (TFBS) motifs with higher frequency than comparable endogenous enhancers while using a more selective motif vocabulary, and we show that enhancer activity is correlated with transcription factor expression at the single-cell level. Finally, we characterize causal features of top enhancers via perturbation experiments and show that enhancers as short as 50 bp can maintain specificity. A record of this paper's transparent peer review process is included in the supplemental information.

How do vascular tissues grow with the rest of the developing organism postnatally? In this issue of Cell Systems, Pi et al. combine modeling with molecular biology to uncover how cell behaviors scale growth. Notably, they discover how proliferation and extrusion regulate endothelial cell number in dorsal aorta development.