Perspective on Oncogenic Processes at the End of the Beginning of Cancer Genomics.

The Cancer Genome Atlas (TCGA) has catalyzed systematic characterization of diverse genomic alterations underlying human cancers. At this historic junction marking the completion of genomic characterization of over 11,000 tumors from 33 cancer types, we present our current understanding of the molecular processes governing oncogenesis. We illustrate our insights into cancer through synthesis of the findings of the TCGA PanCancer Atlas project on three facets of oncogenesis: (1) somatic driver mutations, germline pathogenic variants, and their interactions in the tumor; (2) the influence of the tumor genome and epigenome on transcriptome and proteome; and (3) the relationship between tumor and the microenvironment, including implications for drugs targeting driver events and immunotherapies. These results will anchor future characterization of rare and common tumor types, primary and relapsed tumors, and cancers across ancestry groups and will guide the deployment of clinical genomic sequencing.

Comprehensive Characterization of Cancer Driver Genes and Mutations.

Identifying molecular cancer drivers is critical for precision oncology. Multiple advanced algorithms to identify drivers now exist, but systematic attempts to combine and optimize them on large datasets are few. We report a PanCancer and PanSoftware analysis spanning 9,423 tumor exomes (comprising all 33 of The Cancer Genome Atlas projects) and using 26 computational tools to catalog driver genes and mutations. We identify 299 driver genes with implications regarding their anatomical sites and cancer/cell types. Sequence- and structure-based analyses identified >3,400 putative missense driver mutations supported by multiple lines of evidence. Experimental validation confirmed 60%-85% of predicted mutations as likely drivers. We found that >300 MSI tumors are associated with high PD-1/PD-L1, and 57% of tumors analyzed harbor putative clinically actionable events. Our study represents the most comprehensive discovery of cancer genes and mutations to date and will serve as a blueprint for future biological and clinical endeavors.

Floria: fast and accurate strain haplotyping in metagenomes.

SUMMARY: Shotgun metagenomics allows for direct analysis of microbial community genetics, but scalable computational methods for the recovery of bacterial strain genomes from microbiomes remains a key challenge. We introduce Floria, a novel method designed for rapid and accurate recovery of strain haplotypes from short and long-read metagenome sequencing data, based on minimum error correction (MEC) read clustering and a strain-preserving network flow model. Floria can function as a standalone haplotyping method, outputting alleles and reads that co-occur on the same strain, as well as an end-to-end read-to-assembly pipeline (Floria-PL) for strain-level assembly. Benchmarking evaluations on synthetic metagenomes show that Floria is > 3× faster and recovers 21% more strain content than base-level assembly methods (Strainberry) while being over an order of magnitude faster when only phasing is required. Applying Floria to a set of 109 deeply sequenced nanopore metagenomes took <20 min on average per sample and identified several species that have consistent strain heterogeneity. Applying Floria's short-read haplotyping to a longitudinal gut metagenomics dataset revealed a dynamic multi-strain Anaerostipes hadrus community with frequent strain loss and emergence events over 636 days. With Floria, accurate haplotyping of metagenomic datasets takes mere minutes on standard workstations, paving the way for extensive strain-level metagenomic analyses. AVAILABILITY AND IMPLEMENTATION: Floria is available at https://github.com/bluenote-1577/floria, and the Floria-PL pipeline is available at https://github.com/jsgounot/Floria_analysis_workflow along with code for reproducing the benchmarks.

CREAMMIST: an integrative probabilistic database for cancer drug response prediction.

Extensive in vitro cancer drug screening datasets have enabled scientists to identify biomarkers and develop machine learning models for predicting drug sensitivity. While most advancements have focused on omics profiles, cancer drug sensitivity scores precalculated by the original sources are often used as-is, without consideration for variabilities between studies. It is well-known that significant inconsistencies exist between the drug sensitivity scores across datasets due to differences in experimental setups and preprocessing methods used to obtain the sensitivity scores. As a result, many studies opt to focus only on a single dataset, leading to underutilization of available data and a limited interpretation of cancer pharmacogenomics analysis. To overcome these caveats, we have developed CREAMMIST (https://creammist.mtms.dev), an integrative database that enables users to obtain an integrative dose-response curve, to capture uncertainty (or high certainty when multiple datasets well align) across five widely used cancer cell-line drug-response datasets. We utilized the Bayesian framework to systematically integrate all available dose-response values across datasets (>14 millions dose-response data points). CREAMMIST provides easy-to-use statistics derived from the integrative dose-response curves for various downstream analyses such as identifying biomarkers, selecting drug concentrations for experiments, and training robust machine learning models.

MetageNN: a memory-efficient neural network taxonomic classifier robust to sequencing errors and missing genomes.

BACKGROUND: With the rapid increase in throughput of long-read sequencing technologies, recent studies have explored their potential for taxonomic classification by using alignment-based approaches to reduce the impact of higher sequencing error rates. While alignment-based methods are generally slower, k-mer-based taxonomic classifiers can overcome this limitation, potentially at the expense of lower sensitivity for strains and species that are not in the database. RESULTS: We present MetageNN, a memory-efficient long-read taxonomic classifier that is robust to sequencing errors and missing genomes. MetageNN is a neural network model that uses short k-mer profiles of sequences to reduce the impact of distribution shifts on error-prone long reads. Benchmarking MetageNN against other machine learning approaches for taxonomic classification (GeNet) showed substantial improvements with long-read data (20% improvement in F1 score). By utilizing nanopore sequencing data, MetageNN exhibits improved sensitivity in situations where the reference database is incomplete. It surpasses the alignment-based MetaMaps and MEGAN-LR, as well as the k-mer-based Kraken2 tools, with improvements of 100%, 36%, and 23% respectively at the read-level analysis. Notably, at the community level, MetageNN consistently demonstrated higher sensitivities than the previously mentioned tools. Furthermore, MetageNN requires < 1/4th of the database storage used by Kraken2, MEGAN-LR and MMseqs2 and is > 7× faster than MetaMaps and GeNet and > 2× faster than MEGAN-LR and MMseqs2. CONCLUSION: This proof of concept work demonstrates the utility of machine-learning-based methods for taxonomic classification using long reads. MetageNN can be used on sequences not classified by conventional methods and offers an alternative approach for memory-efficient classifiers that can be optimized further.

Comparative analysis of metagenomic classifiers for long-read sequencing datasets.

BACKGROUND: Long reads have gained popularity in the analysis of metagenomics data. Therefore, we comprehensively assessed metagenomics classification tools on the species taxonomic level. We analysed kmer-based tools, mapping-based tools and two general-purpose long reads mappers. We evaluated more than 20 pipelines which use either nucleotide or protein databases and selected 13 for an extensive benchmark. We prepared seven synthetic datasets to test various scenarios, including the presence of a host, unknown species and related species. Moreover, we used available sequencing data from three well-defined mock communities, including a dataset with abundance varying from 0.0001 to 20% and six real gut microbiomes. RESULTS: General-purpose mappers Minimap2 and Ram achieved similar or better accuracy on most testing metrics than best-performing classification tools. They were up to ten times slower than the fastest kmer-based tools requiring up to four times less RAM. All tested tools were prone to report organisms not present in datasets, except CLARK-S, and they underperformed in the case of the high presence of the host's genetic material. Tools which use a protein database performed worse than those based on a nucleotide database. Longer read lengths made classification easier, but due to the difference in read length distributions among species, the usage of only the longest reads reduced the accuracy. The comparison of real gut microbiome datasets shows a similar abundance profiles for the same type of tools but discordance in the number of reported organisms and abundances between types. Most assessments showed the influence of database completeness on the reports. CONCLUSION: The findings indicate that kmer-based tools are well-suited for rapid analysis of long reads data. However, when heightened accuracy is essential, mappers demonstrate slightly superior performance, albeit at a considerably slower pace. Nevertheless, a combination of diverse categories of tools and databases will likely be necessary to analyse complex samples. Discrepancies observed among tools when applied to real gut datasets, as well as a reduced performance in cases where unknown species or a significant proportion of the host genome is present in the sample, highlight the need for continuous improvement of existing tools. Additionally, regular updates and curation of databases are important to ensure their effectiveness.

The skin microbiome in pediatric atopic dermatitis and food allergy.

The skin microbiome is an extensive community of bacteria, fungi, mites, viruses and archaea colonizing the skin. Fluctuations in the composition of the skin microbiome have been observed in atopic dermatitis (AD) and food allergy (FA), particularly in early life, established disease, and associated with therapeutics. However, AD is a multifactorial disease characterized by skin barrier aberrations modulated by genetics, immunology, and environmental influences, thus the skin microbiome is not the sole feature of this disease. Future research should focus on mechanistic understanding of how early-life skin microbial shifts may influence AD and FA onset, to guide potential early intervention strategies or as microbial biomarkers to identify high-risk infants who may benefit from possible microbiome-based biotherapeutic strategies. Harnessing skin microbes as AD biotherapeutics is an emerging field, but more work is needed to investigate whether this approach can lead to sustained clinical responses.

In Vivo Mapping of Eukaryotic RNA Interactomes Reveals Principles of Higher-Order Organization and Regulation.

Identifying pairwise RNA-RNA interactions is key to understanding how RNAs fold and interact with other RNAs inside the cell. We present a high-throughput approach, sequencing of psoralen crosslinked, ligated, and selected hybrids (SPLASH), that maps pairwise RNA interactions in vivo with high sensitivity and specificity, genome-wide. Applying SPLASH to human and yeast transcriptomes revealed the diversity and dynamics of thousands of long-range intra- and intermolecular RNA-RNA interactions. Our analysis highlighted key structural features of RNA classes, including the modular organization of mRNAs, its impact on translation and decay, and the enrichment of long-range interactions in noncoding RNAs. Additionally, intermolecular mRNA interactions were organized into network clusters and were remodeled during cellular differentiation. We also identified hundreds of known and new snoRNA-rRNA binding sites, expanding our knowledge of rRNA biogenesis. These results highlight the underexplored complexity of RNA interactomes and pave the way to better understanding how RNA organization impacts biology.

Shared signatures and divergence in skin microbiomes of children with atopic dermatitis and their caregivers.

BACKGROUND: Atopic dermatitis (AD) is a common chronic skin condition in children (15-20%) that can significantly impair their quality of life. As a result of its relapsing nature and enrichment of Staphylococcus aureus during flares, clinical management can include eradicating S aureus from the skin of children; however, this does not extend to their healthy caregivers, who are potential reservoirs. OBJECTIVE: Our aim was to understand skin microbiome sharing and microbial features in children with AD and their healthy adult caregivers. METHODS: We utilized whole-metagenome profiling at 4 body sites (volar forearm, antecubital fossae, cheeks, and lesions) in combination with sequencing of S aureus isolates to characterize a cohort of children with AD and their healthy caregivers (n = 30 families) compared to matched pairs from control households (n = 30 families). RESULTS: Metagenomic analysis revealed distinct microbiome configurations in the nonlesional skin of AD children and their healthy caregivers versus controls, which were sufficient to accurately predict case-control status (area under the receiver operating characteristic curve > 0.8). These differences were accompanied by significant microbiome similarity between children and their caregivers, indicating that microbiome sharing may play a role in recurrent disease flares. Whole-genome comparisons with high-quality S aureus isolate genomes (n = 55) confirmed significant strain sharing between AD children and their caregivers and AD-specific enrichment of strains expressing enterotoxins Q and K/K2. CONCLUSION: Our results highlight the distinctive skin microbiome features of healthy caregivers for children with AD and support their inclusion in strategies for the treatment of recurrent pediatric AD.

Profiling Microbial Communities in Idiopathic Granulomatous Mastitis.

Idiopathic granulomatous mastitis (IGM) is a rare and benign inflammatory breast disease with ambiguous aetiology. Contrastingly, lactational mastitis (LM) is commonly diagnosed in breastfeeding women. To investigate IGM aetiology, we profiled the microbial flora of pus and skin in patients with IGM and LM. A total of 26 patients with IGM and 6 patients with LM were included in the study. The 16S rRNA sequencing libraries were constructed from 16S rRNA gene amplified from total DNA extracted from pus and skin swabs in patients with IGM and LM controls. Constructed libraries were multiplexed and paired-end sequenced on HiSeq4000. Metagenomic analysis was conducted using modified microbiome abundance analysis suite customised R-resource for paired pus and skin samples. Microbiome multivariable association analyses were performed using linear models. A total of 21 IGM and 3 LM paired pus and skin samples underwent metagenomic analysis. Bray−Curtis ecological dissimilarity distance showed dissimilarity across four sample types (IGM pus, IGM skin, LM pus, and LM skin; PERMANOVA, p < 0.001). No characteristic dominant genus was observed across the IGM samples. The IGM pus samples were more diverse than corresponding IGM skin samples (Shannon and Simpson index; Wilcoxon paired signed-rank tests, p = 0.022 and p = 0.07). Corynebacterium kroppenstedtii, reportedly associated with IGM in the literature, was higher in IGM pus samples than paired skin samples (Wilcoxon, p = 0.022). Three other species and nineteen genera were statistically significant in paired IGM pus−skin comparison after antibiotic treatment adjustment and multiple comparisons correction. Microbial profiles are unique between patients with IGM and LM. Inter-patient variability and polymicrobial IGM pus samples cannot implicate specific genus or species as an infectious cause for IGM.

Dominant Carbapenemase-Encoding Plasmids in Clinical Enterobacterales Isolates and Hypervirulent Klebsiella pneumoniae, Singapore.

Dissemination of carbapenemase-encoding plasmids by horizontal gene transfer in multidrug-resistant bacteria is the major driver of rising carbapenem-resistance, but the conjugative mechanics and evolution of clinically relevant plasmids are not yet clear. We performed whole-genome sequencing on 1,215 clinical Enterobacterales isolates collected in Singapore during 2010-2015. We identified 1,126 carbapenemase-encoding plasmids and discovered pKPC2 is becoming the dominant plasmid in Singapore, overtaking an earlier dominant plasmid, pNDM1. pKPC2 frequently conjugates with many Enterobacterales species, including hypervirulent Klebsiella pneumoniae, and maintains stability in vitro without selection pressure and minimal adaptive sequence changes. Furthermore, capsule and decreasing taxonomic relatedness between donor and recipient pairs are greater conjugation barriers for pNDM1 than pKPC2. The low fitness costs pKPC2 exerts in Enterobacterales species indicate previously undetected carriage selection in other ecological settings. The ease of conjugation and stability of pKPC2 in hypervirulent K. pneumoniae could fuel spread into the community.

TUGDA: task uncertainty guided domain adaptation for robust generalization of cancer drug response prediction from in vitro to in vivo settings.

MOTIVATION: Large-scale cancer omics studies have highlighted the diversity of patient molecular profiles and the importance of leveraging this information to deliver the right drug to the right patient at the right time. Key challenges in learning predictive models for this include the high-dimensionality of omics data and heterogeneity in biological and clinical factors affecting patient response. The use of multi-task learning techniques has been widely explored to address dataset limitations for in vitro drug response models, while domain adaptation (DA) has been employed to extend them to predict in vivo response. In both of these transfer learning settings, noisy data for some tasks (or domains) can substantially reduce the performance for others compared to single-task (domain) learners, i.e. lead to negative transfer (NT). RESULTS: We describe a novel multi-task unsupervised DA method (TUGDA) that addresses these limitations in a unified framework by quantifying uncertainty in predictors and weighting their influence on shared feature representations. TUGDA's ability to rely more on predictors with low-uncertainty allowed it to notably reduce cases of NT for in vitro models (94% overall) compared to state-of-the-art methods. For DA to in vivo settings, TUGDA improved over previous methods for patient-derived xenografts (9 out of 14 drugs) as well as patient datasets (significant associations in 9 out of 22 drugs). TUGDA's ability to avoid NT thus provides a key capability as we try to integrate diverse drug-response datasets to build consistent predictive models with in vivo utility. AVAILABILITYAND IMPLEMENTATION: https://github.com/CSB5/TUGDA. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Boosting natural history research via metagenomic clean-up of crowdsourced feces.

Biodiversity is in crisis due to habitat destruction and climate change. The conservation of many noncharismatic species is hampered by the lack of data. Yet, natural history research-a major source of information on noncharismatic species-is in decline. We here suggest a remedy for many mammal species, i.e., metagenomic clean-up of fecal samples that are "crowdsourced" during routine field surveys. Based on literature data, we estimate that this approach could yield natural history information for circa 1,000 species within a decade. Metagenomic analysis would simultaneously yield natural history data on diet and gut parasites while enhancing our understanding of host genetics, gut microbiome, and the functional interactions between traditional and new natural history data. We document the power of this approach by carrying out a "metagenomic clean-up" on fecal samples collected during a single night of small mammal trapping in one of Alfred Wallace's favorite collecting sites.

BEEM-Static: Accurate inference of ecological interactions from cross-sectional microbiome data.

CONCLUSION: BEEM-Static provides new opportunities for mining ecologically interpretable interactions and systems insights from the growing corpus of microbiome data.

Atopic dermatitis microbiomes stratify into ecologic dermotypes enabling microbial virulence and disease severity.

BACKGROUND: Atopic dermatitis (AD) is a common skin disease affecting up to 20% of the global population, with significant clinical heterogeneity and limited information about molecular subtypes and actionable biomarkers. Although alterations in the skin microbiome have been described in subjects with AD during progression to flare state, the prognostic value of baseline microbiome configurations has not been explored. OBJECTIVE: Our aim was to identify microbial signatures on AD skin that are predictive of disease fate. METHODS: Nonlesional skin of patients with AD and healthy control subjects were sampled at 2 time points separated by at least 4 weeks. Using whole metagenome analysis of skin microbiomes of patients with AD and control subjects (n = 49 and 189 samples), we identified distinct microbiome configurations (dermotypes A and B). Blood was collected for immunophenotyping, and skin surface samples were analyzed for correlations with natural moisturizing factors and antimicrobial peptides. RESULTS: Dermotypes were robust and validated across 2 additional cohorts (63 individuals), with strong enrichment of subjects with AD in dermotype B. Dermotype B was characterized by reduced microbial richness, depletion of Cutibacterium acnes, Dermacoccus and Methylobacterium species, individual-specific outlier abundance of Staphylococcus species (eg, S epidermidis, S capitis, S aureus), and enrichment in metabolic pathways (eg, branched chain amino acids and arginine biosynthesis) and virulence genes (eg, ß-toxin, δ-toxin) that defined a pathogenic ecology. Skin surface and circulating host biomarkers exhibited a distinct microbial-associated signature that was further reflected in more severe itching, frequent flares, and increased disease severity in patients harboring the dermotype B microbiome. CONCLUSION: We report distinct clusters of microbial profiles that delineate the role of microbiome configurations in AD heterogeneity, highlight a mechanism for ongoing inflammation, and provide prognostic utility toward microbiome-based disease stratification.

Evaluating the accuracy of Listeria monocytogenes assemblies from quasimetagenomic samples using long and short reads.

BACKGROUND: Whole genome sequencing of cultured pathogens is the state of the art public health response for the bioinformatic source tracking of illness outbreaks. Quasimetagenomics can substantially reduce the amount of culturing needed before a high quality genome can be recovered. Highly accurate short read data is analyzed for single nucleotide polymorphisms and multi-locus sequence types to differentiate strains but cannot span many genomic repeats, resulting in highly fragmented assemblies. Long reads can span repeats, resulting in much more contiguous assemblies, but have lower accuracy than short reads. RESULTS: We evaluated the accuracy of Listeria monocytogenes assemblies from enrichments (quasimetagenomes) of naturally-contaminated ice cream using long read (Oxford Nanopore) and short read (Illumina) sequencing data. Accuracy of ten assembly approaches, over a range of sequencing depths, was evaluated by comparing sequence similarity of genes in assemblies to a complete reference genome. Long read assemblies reconstructed a circularized genome as well as a 71 kbp plasmid after 24 h of enrichment; however, high error rates prevented high fidelity gene assembly, even at 150X depth of coverage. Short read assemblies accurately reconstructed the core genes after 28 h of enrichment but produced highly fragmented genomes. Hybrid approaches demonstrated promising results but had biases based upon the initial assembly strategy. Short read assemblies scaffolded with long reads accurately assembled the core genes after just 24 h of enrichment, but were highly fragmented. Long read assemblies polished with short reads reconstructed a circularized genome and plasmid and assembled all the genes after 24 h enrichment but with less fidelity for the core genes than the short read assemblies. CONCLUSION: The integration of long and short read sequencing of quasimetagenomes expedited the reconstruction of a high quality pathogen genome compared to either platform alone. A new and more complete level of information about genome structure, gene order and mobile elements can be added to the public health response by incorporating long read analyses with the standard short read WGS outbreak response.

Household transmission of carbapenemase-producing Enterobacteriaceae: a prospective cohort study.

OBJECTIVES: To estimate the transmission rate of carbapenemase-producing Enterobacteriaceae (CPE) in households with recently hospitalized CPE carriers. METHODS: We conducted a prospective case-ascertained cohort study. We identified the presence of CPE in stool samples from index subjects, household contacts and companion animals and environmental samples at regular intervals. Linked transmissions were identified by WGS. A Markov model was constructed to estimate the household transmission potential of CPE. RESULTS: Ten recently hospitalized index patients and 14 household contacts were included. There were seven households with one contact, two households with two contacts, and one household with three contacts. Index patients were colonized with blaOXA-48-like (n = 4), blaKPC-2 (n = 3), blaIMP (n = 2), and blaNDM-1 (n = 1), distributed among divergent species of Enterobacteriaceae. After a cumulative follow-up time of 9.0 years, three family members (21.4%, 3/14) acquired four different types of CPE in the community (hazard rate of 0.22/year). The probability of CPE transmission from an index patient to a household contact was 10% (95% CI 4%-26%). CONCLUSIONS: We observed limited transmission of CPE from an index patient to household contacts. Larger studies are needed to understand the factors associated with household transmission of CPE and identify preventive strategies.

Extended-Spectrum ß-Lactamase-Producing and mcr-1-Positive Escherichia coli from the Gut Microbiota of Healthy Singaporeans.

Multidrug-resistant (MDR) Escherichia coli strains that carry extended-spectrum ß-lactamases (ESBLs) or colistin resistance gene mcr-1 have been identified in the human gut at an increasing incidence worldwide. In this study, we isolated and characterized MDR Enterobacteriaceae from the gut microbiota of healthy Singaporeans and show that the detection rates for ESBL-producing and mcr-positive Enterobacteriaceae are 25.7% (28/109) and 7.3% (8/109), respectively. Whole-genome sequencing analysis of the 37 E. coli isolates assigned them into 25 sequence types and 6 different phylogroups, suggesting that the MDR E. coli gut colonizers are highly diverse. We then analyzed the genetic context of the resistance genes and found that composite transposons played important roles in the cotransfer of blaCTX-M-15/55 and qnrS1, as well as the acquisition of mcr-1. Furthermore, comparative genomic analysis showed that 12 of the 37 MDR E. coli isolates showed high similarity to ESBL-producing E. coli isolates from raw meat products in local markets. By analyzing the core genome single nucleotide polymorphisms (SNPs) shared by these isolates, we identified possible clonal transmission of an MDR E. coli clone between human and raw meat, as well as a group of highly similar IncI2 (Delta) plasmids that might be responsible for the dissemination of mcr-1 in a much wider geographic region. Together, these results suggest that antibiotic resistance may be transmitted between different environmental settings by the expansion of MDR E. coli clones, as well as by the dissemination of resistance plasmids. IMPORTANCE The human gut can harbor both antibiotic-resistant and virulent Escherichia coli which may subsequently cause infections. In this study, we found that multidrug-resistant (MDR) E. coli isolates from the gut of healthy Singaporeans carry a diverse range of antibiotic resistance mechanisms and virulence factor genes and are highly diverse. By comparing their genomes with the extended-spectrum ß-lactamase (ESBL)-producing E. coli isolates from raw meat products that were sampled at a similar time from local markets, we detected an MDR E. coli clone that was possibly transmitted between humans and raw meat products. Furthermore, we also found that a group of resistance plasmids might be responsible for the dissemination of colistin resistance gene mcr-1 in Singapore, Malaysia, and Europe. Our findings call for better countermeasures to block the transmission of antibiotic resistance.

Cloning and variation of ground state intestinal stem cells.

Stem cells of the gastrointestinal tract, pancreas, liver and other columnar epithelia collectively resist cloning in their elemental states. Here we demonstrate the cloning and propagation of highly clonogenic, 'ground state' stem cells of the human intestine and colon. We show that derived stem-cell pedigrees sustain limited copy number and sequence variation despite extensive serial passaging and display exquisitely precise, cell-autonomous commitment to epithelial differentiation consistent with their origins along the intestinal tract. This developmentally patterned and epigenetically maintained commitment of stem cells is likely to enforce the functional specificity of the adult intestinal tract. Using clonally derived colonic epithelia, we show that toxins A or B of the enteric pathogen Clostridium difficile recapitulate the salient features of pseudomembranous colitis. The stability of the epigenetic commitment programs of these stem cells, coupled with their unlimited replicative expansion and maintained clonogenicity, suggests certain advantages for their use in disease modelling and regenerative medicine.