Disrupting cellular memory to overcome drug resistance.

Gene expression states persist for varying lengths of time at the single-cell level, a phenomenon known as gene expression memory. When cells switch states, losing memory of their prior state, this transition can occur in the absence of genetic changes. However, we lack robust methods to find regulators of memory or track state switching. Here, we develop a lineage tracing-based technique to quantify memory and identify cells that switch states. Applied to melanoma cells without therapy, we quantify long-lived fluctuations in gene expression that are predictive of later resistance to targeted therapy. We also identify the PI3K and TGF-ß pathways as state switching modulators. We propose a pretreatment model, first applying a PI3K inhibitor to modulate gene expression states, then applying targeted therapy, which leads to less resistance than targeted therapy alone. Together, we present a method for finding modulators of gene expression memory and their associated cell fates.

Benchmarking Gene Regulatory Network Inference Methods on Simulated and Experimental Data.

Although the challenge of gene regulatory network inference has been studied for more than a decade, it is still unclear how well network inference methods work when applied to real data. Attempts to benchmark these methods on experimental data have yielded mixed results, in which sometimes even the best methods fail to outperform random guessing, and in other cases they perform reasonably well. So, one of the most valuable contributions one can currently make to the field of network inference is to benchmark methods on experimental data for which the true underlying network is already known, and report the results so that we can get a clearer picture of their efficacy. In this paper, we report results from the first, to our knowledge, benchmarking of network inference methods on single cell E. coli transcriptomic data. We report a moderate level of accuracy for the methods, better than random chance but still far from perfect. We also find that some methods that were quite strong and accurate on microarray and bulk RNA-seq data did not perform as well on the single cell data. Additionally, we benchmark a simple network inference method (Pearson correlation), on data generated through computer simulations in order to draw conclusions about general best practices in network inference studies. We predict that network inference would be more accurate using proteomic data rather than transcriptomic data, which could become relevant if high-throughput proteomic experimental methods are developed in the future. We also show through simulations that using a simplified model of gene expression that skips the mRNA step tends to substantially overestimate the accuracy of network inference methods, and advise against using this model for future in silico benchmarking studies.

Epigenomic Blood-Based Early Detection of Pancreatic Cancer Employing Cell-Free DNA.

BACKGROUND & AIMS: Early detection of pancreatic cancer (PaC) can drastically improve survival rates. Approximately 25% of subjects with PaC have type 2 diabetes diagnosed within 3 years prior to the PaC diagnosis, suggesting that subjects with type 2 diabetes are at high risk of occult PaC. We have developed an early-detection PaC test, based on changes in 5-hydroxymethylcytosine (5hmC) signals in cell-free DNA from plasma. METHODS: Blood was collected from 132 subjects with PaC and 528 noncancer subjects to generate epigenomic and genomic feature sets yielding a predictive PaC signal algorithm. The algorithm was validated in a blinded cohort composed of 102 subjects with PaC, 2048 noncancer subjects, and 1524 subjects with non-PaCs. RESULTS: 5hmC differential profiling and additional genomic features enabled the development of a machine learning algorithm capable of distinguishing subjects with PaC from noncancer subjects with high specificity and sensitivity. The algorithm was validated with a sensitivity for early-stage (stage I/II) PaC of 68.3% (95% confidence interval [CI], 51.9%-81.9%) and an overall specificity of 96.9% (95% CI, 96.1%-97.7%). CONCLUSIONS: The PaC detection test showed robust early-stage detection of PaC signal in the studied cohorts with varying type 2 diabetes status. This assay merits further clinical validation for the early detection of PaC in high-risk individuals.

A scalable, open-source implementation of a large-scale mechanistic model for single cell proliferation and death signaling.

Mechanistic models of how single cells respond to different perturbations can help integrate disparate big data sets or predict response to varied drug combinations. However, the construction and simulation of such models have proved challenging. Here, we developed a python-based model creation and simulation pipeline that converts a few structured text files into an SBML standard and is high-performance- and cloud-computing ready. We applied this pipeline to our large-scale, mechanistic pan-cancer signaling model (named SPARCED) and demonstrate it by adding an IFNγ pathway submodel. We then investigated whether a putative crosstalk mechanism could be consistent with experimental observations from the LINCS MCF10A Data Cube that IFNγ acts as an anti-proliferative factor. The analyses suggested this observation can be explained by IFNγ-induced SOCS1 sequestering activated EGF receptors. This work forms a foundational recipe for increased mechanistic model-based data integration on a single-cell level, an important building block for clinically-predictive mechanistic models.

Probing transient memory of cellular states using single-cell lineages.

The inherent stochasticity in the gene product levels can drive single cells within an isoclonal population to different phenotypic states. The dynamic nature of this intercellular variation, where individual cells can transition between different states over time, makes it a particularly hard phenomenon to characterize. We reviewed recent progress in leveraging the classical Luria-Delbrück experiment to infer the transient heritability of the cellular states. Similar to the original experiment, individual cells were first grown into cell colonies, and then, the fraction of cells residing in different states was assayed for each colony. We discuss modeling approaches for capturing dynamic state transitions in a growing cell population and highlight formulas that identify the kinetics of state switching from the extent of colony-to-colony fluctuations. The utility of this method in identifying multi-generational memory of the both expression and phenotypic states is illustrated across diverse biological systems from cancer drug resistance, reactivation of human viruses, and cellular immune responses. In summary, this fluctuation-based methodology provides a powerful approach for elucidating cell-state transitions from a single time point measurement, which is particularly relevant in situations where measurements lead to cell death (as in single-cell RNA-seq or drug treatment) or cause an irreversible change in cell physiology.

Insights on the Control of Yeast Single-Cell Growth Variability by Members of the Trehalose Phosphate Synthase (TPS) Complex.

Single-cell variability of growth is a biological phenomenon that has attracted growing interest in recent years. Important progress has been made in the knowledge of the origin of cell-to-cell heterogeneity of growth, especially in microbial cells. To better understand the origins of such heterogeneity at the single-cell level, we developed a new methodological pipeline that coupled cytometry-based cell sorting with automatized microscopy and image analysis to score the growth rate of thousands of single cells. This allowed investigating the influence of the initial amount of proteins of interest on the subsequent growth of the microcolony. As a preliminary step to validate this experimental setup, we referred to previous findings in yeast where the expression level of Tsl1, a member of the Trehalose Phosphate Synthase (TPS) complex, negatively correlated with cell division rate. We unfortunately could not find any influence of the initial TSL1 expression level on the growth rate of the microcolonies. We also analyzed the effect of the natural variations of trehalose-6-phosphate synthase (TPS1) expression on cell-to-cell growth heterogeneity, but we did not find any correlation. However, due to the already known altered growth of the tps1Δ mutants, we tested this strain at the single-cell level on a permissive carbon source. This mutant showed an outstanding lack of reproducibility of growth rate distributions as compared to the wild-type strain, with variable proportions of non-growing cells between cultivations and more heterogeneous microcolonies in terms of individual growth rates. Interestingly, this variable behavior at the single-cell level was reminiscent to the high variability that is also stochastically suffered at the population level when cultivating this tps1Δ strain, even when using controlled bioreactors.

Detection of early stage pancreatic cancer using 5-hydroxymethylcytosine signatures in circulating cell free DNA.

Pancreatic cancer is often detected late, when curative therapies are no longer possible. Here, we present non-invasive detection of pancreatic ductal adenocarcinoma (PDAC) by 5-hydroxymethylcytosine (5hmC) changes in circulating cell free DNA from a PDAC cohort (n = 64) in comparison with a non-cancer cohort (n = 243). Differential hydroxymethylation is found in thousands of genes, most significantly in genes related to pancreas development or function (GATA4, GATA6, PROX1, ONECUT1, MEIS2), and cancer pathogenesis (YAP1, TEAD1, PROX1, IGF1). cfDNA hydroxymethylome in PDAC cohort is differentially enriched for genes that are commonly de-regulated in PDAC tumors upon activation of KRAS and inactivation of TP53. Regularized regression models built using 5hmC densities in genes perform with AUC of 0.92 (discovery dataset, n = 79) and 0.92-0.94 (two independent test sets, n = 228). Furthermore, tissue-derived 5hmC features can be used to classify PDAC cfDNA (AUC = 0.88). These findings suggest that 5hmC changes enable classification of PDAC even during early stage disease.

Gene Networks with Transcriptional Bursting Recapitulate Rare Transient Coordinated High Expression States in Cancer.

Non-genetic transcriptional variability is a potential mechanism for therapy resistance in melanoma. Specifically, rare subpopulations of cells occupy a transient pre-resistant state characterized by coordinated high expression of several genes and survive therapy. How might these rare states arise and disappear within the population? It is unclear whether the canonical models of probabilistic transcriptional pulsing can explain this behavior, or if it requires special, hitherto unidentified mechanisms. We show that a minimal model of transcriptional bursting and gene interactions can give rise to rare coordinated high expression states. These states occur more frequently in networks with low connectivity and depend on three parameters. While entry into these states is initiated by a long transcriptional burst that also triggers entry of other genes, the exit occurs through independent inactivation of individual genes. Together, we demonstrate that established principles of gene regulation are sufficient to describe this behavior and argue for its more general existence. A record of this paper's transparent peer review process is included in the Supplemental Information.

Network inference in systems biology: recent developments, challenges, and applications.

One of the most interesting, difficult, and potentially useful topics in computational biology is the inference of gene regulatory networks (GRNs) from expression data. Although researchers have been working on this topic for more than a decade and much progress has been made, it remains an unsolved problem and even the most sophisticated inference algorithms are far from perfect. In this paper, we review the latest developments in network inference, including state-of-the-art algorithms like PIDC, Phixer, and more. We also discuss unsolved computational challenges, including the optimal combination of algorithms, integration of multiple data sources, and pseudo-temporal ordering of static expression data. Lastly, we discuss some exciting applications of network inference in cancer research, and provide a list of useful software tools for researchers hoping to conduct their own network inference analyses.

Antiproliferative and erythroid differentiation of piperazine and triphenyl derivatives against k-562 human chronic myelogenous leukemia.

Five piperazine derivatives (S)-4-benzyl-1-(4-bromo-3-methylphenyl)-2 methylpiperazine (A), (S)-1-benzyl-3-isobutylpiperazine-2,5-dione (B), (S)-1-benzyl-3 methylpiperazine-2,5-dione (C), (S)-1,3-dibenzylpiperazine-2,5-dione (D), (E)-1-(3-methyl 4-((E)-3-(2-methylpropylidene) piperazin-1-yl) phenyl)-2-(2 methylpropylidene) piperazine (E) and triphenyl derivative ammonium 2-((2,3',3''-trimethyl-[1,1':4',1''-terphenyl]-4 yl)oxy)acetate (F) were tested for inhibition of K-562 cell proliferation and for induction of erythroid differentiation. Among them, two piperazine and one triphenyl derivatives, compounds A, E, and F inhibited the proliferation of the K562 cell lines exhibiting inhibition concentration 50 (IC50) (IC50) of values 30.10±1.6, 4.60±0.4 and 25.70±1.10 µg ml(-1), respectively. If compound A and F were added to suboptimal concentrations of the established anticancer drugs cytosine arabinoside or mithramycin, pronounced synergic effects were observed.