Boris Adryan

Many genes are expressed in bursts, which can contribute to cell-to-cell heterogeneity. It is now possible to measure this heterogeneity with high throughput single cell gene expression assays (single cell qPCR and RNA-seq). These experimental approaches generate gene expression distributions which can be used to estimate the kinetic parameters of gene expression bursting, namely the rate that genes turn on, the rate that genes turn off, and the rate of transcription. We construct a complete pipeline for the analysis of single cell qPCR data that uses the mathematics behind bursty expression to develop more accurate and robust algorithms for analyzing the origin of heterogeneity in experimental samples, specifically an algorithm for clustering cells by their bursting behavior (Simulated Annealing for Bursty Expression Clustering, SABEC) and a statistical tool for comparing the kinetic parameters of bursty expression across populations of cells (Estimation of Parameter changes in Kinetics, EPiK). We applied these methods to hematopoiesis, including a new single cell dataset in which transcription factors (TFs) involved in the earliest branchpoint of blood differentiation were individually up- and down-regulated. We could identify two unique sub-populations within a seemingly homogenous group of hematopoietic stem cells. In addition, we could predict regulatory mechanisms controlling the expression levels of eighteen key hematopoietic transcription factors throughout differentiation. Detailed information about gene regulatory mechanisms can therefore be obtained simply from high throughput single cell gene expression data, which should be widely applicable given the rapid expansion of single cell genomics. Many genes are expressed in bursts, which can contribute to cell-to-cell variability. We construct a pipeline for analyzing single cell gene expression data that uses the mathematics behind bursty expression. This pipeline includes one algorithm for clustering cells (Simulated Annealing for Bursty Expression Clustering, SABEC) and a statistical tool for comparing the kinetic parameters of bursty expression across populations of cells (Estimation of Parameter changes in Kinetics, EPiK). We applied these methods to blood development, including a new single cell dataset in which TFs involved in the earliest branchpoint of blood differentiation were individually up- and down-regulated.

Scoring DNA sequences against Position Weight Matrices (PWMs) is a widely adopted method to identify putative transcription factor binding sites. While common bioinformatics tools produce scores that can reflect the binding strength between a specific transcription factor and the DNA, these scores are not directly comparable between different transcription factors. Other methods, including p-value associated approaches (Touzet H, Varré J-S. Efficient and accurate p-value computation for position weight matrices. Algorithms Mol Biol. 2007;2(1510.1186):1748–7188), provide more rigorous ways to identify potential binding sites, but their results are difficult to interpret in terms of binding energy, which is essential for the modeling of transcription factor binding dynamics and enhancer activities. Here, we provide two different ways to find the scaling parameter λ that allows us to infer binding energy from a PWM score. The first approach uses a PWM and background genomic sequence as input to estimate λ for a specific transcription factor, which we applied to show that λ distributions for different transcription factor families correspond with their DNA binding properties. Our second method can reliably convert λ between different PWMs of the same transcription factor, which allows us to directly compare PWMs that were generated by different approaches. These two approaches provide computationally efficient ways to scale PWM scores and estimate the strength of transcription factor binding sites in quantitative studies of binding dynamics. Their results are consistent with each other and previous reports in most of cases.

The binding of transcription factors (TFs) is essential for gene expression. One important characteristic is the actual occupancy of a putative binding site in the genome. In this study, we propose an analytical model to predict genomic occupancy that incorporates the preferred target sequence of a TF in the form of a position weight matrix (PWM), DNA accessibility data (in the case of eukaryotes), the number of TF molecules expected to be bound specifically to the DNA and a parameter that modulates the specificity of the TF. Given actual occupancy data in the form of ChIP-seq profiles, we backwards inferred copy number and specificity for five Drosophila TFs during early embryonic development: Bicoid, Caudal, Giant, Hunchback and Kruppel. Our results suggest that these TFs display thousands of molecules that are specifically bound to the DNA and that whilst Bicoid and Caudal display a higher specificity, the other three TFs (Giant, Hunchback and Kruppel) display lower specificity in their binding (despite having PWMs with higher information content). This study gives further weight to earlier investigations into TF copy numbers that suggest a significant proportion of molecules are not bound specifically to the DNA.

The binding of transcription factors (TFs) is essential for gene expression. One important characteristic is the actual occupancy of a putative binding site in the genome. In this study, we propose an analytical model to predict genomic occupancy that incorporates the preferred target sequence of a TF in the form of a position weight matrix (PWM), DNA accessibility data (in case of eukaryotes), the number of TF molecules expected to be bound specifically to the DNA and a parameter that modulates the specificity of the TF. Given actual occupancy data in form of ChIP-seq profiles, we backwards inferred copy number and specificity for five Drosophila TFs during early embryonic development: Bicoid, Caudal, Giant, Hunchback and Kruppel. Our results suggest that these TFs display thousands of molecules that are specifically bound to the DNA and that, while Bicoid and Caudal display a higher specificity, the other three transcription factors (Giant, Hunchback and Kruppel) display lower specificity in their binding (despite having PWMs with higher information content). This study gives further weight to earlier investigations into TF copy numbers that suggest a significant proportion of molecules are not bound specifically to the DNA.

Regulation of transcription is fundamental to development and physiology, and occurs through binding of transcription factors to specific DNA sequences in the genome. CSL (CBF1/Suppressor of Hairless/LAG-1), a core component of the Notch signaling pathway, is one such transcription factor that acts in concert with co-activators or co-repressors to control the activity of associated target genes. One fundamental question is how CSL can recognize and select among different DNA sequences available in vivo and whether variations between selected sequences can influence its function. We have therefore investigated CSL–DNA recognition using computational approaches to analyze the energetics of CSL bound to different DNAs and tested the in silico predictions with in vitro and in vivo assays. Our results reveal novel aspects of CSL binding that may help explain the range of binding observed in vivo. In addition, using molecular dynamics simulations, we show that domain–domain correlations within CSL differ significantly depending on the DNA sequence bound, suggesting that different DNA sequences may directly influence CSL function. Taken together, our results, based on computational chemistry approaches, provide valuable insights into transcription factor-DNA binding, in this particular case increasing our understanding of CSL–DNA interactions and how these may impact on its transcriptional control.

The organization of binding sites in cis-regulatory elements (CREs) can influence gene expression through a combination of physical mechanisms, ranging from direct interactions between TF molecules to DNA looping and transient chromatin interactions. The study of simple and common building blocks in promoters and other CREs allows us to dissect how all of these mechanisms work together. Many adjacent TF binding sites for the same TF species form homotypic clusters, and these CRE architecture building blocks serve as a prime candidate for understanding interacting transcriptional mechanisms. Homotypic clusters are prevalent in both bacterial and eukaryotic genomes, and are present in both promoters as well as more distal enhancer/silencer elements. Here, we review previous theoretical and experimental studies that show how the complexity (number of binding sites) and spatial organization (distance between sites and overall distance from transcription start sites) of homotypic clusters influence gene expression. In particular, we describe how homotypic clusters modulate the temporal dynamics of TF binding, a mechanism that can affect gene expression, but which has not yet been sufficiently characterized. We propose further experiments on homotypic clusters that would be useful in developing mechanistic models of gene expression.

Site-specific transcription factors (TFs) bind to their target sites on the DNA, where they regulate the rate at which genes are transcribed. Bacterial TFs undergo facilitated diffusion (a combination of 3D diffusion around and 1D random walk on the DNA) when searching for their target sites. Using computer simulations of this search process, we show that the organization of the binding sites, in conjunction with TF copy number and binding site affinity, plays an important role in determining not only the steady state of promoter occupancy, but also the order at which TFs bind. These effects can be captured by facilitated diffusion-based models, but not by standard thermodynamics. We show that the spacing of binding sites encodes complex logic, which can be derived from combinations of three basic building blocks: switches, barriers and clusters, whose response alone and in higher orders of organization we characterize in detail. Effective promoter organizations are commonly found in the E. coli genome and are highly conserved between strains. This will allow studies of gene regulation at a previously unprecedented level of detail, where our framework can create testable hypothesis of promoter logic.

Site-specific transcription factors (TFs) bind to their target sites on the DNA, where they regulate the rate at which genes are transcribed. Bacterial TFs undergo facilitated diffusion (a combination of 3D diffusion around and 1D random walk on the DNA) when searching for their target sites. Using computer simulations of this search process, we show that the organisation of the binding sites, in conjunction with TF copy number and binding site affinity, plays an important role in determining not only the steady state of promoter occupancy, but also the order at which TFs bind. These effects can be captured by facilitated diffusion-based models, but not by standard thermodynamics. We show that the spacing of binding sites encodes complex logic, which can be derived from combinations of three basic building blocks: switches, barriers and clusters, whose response alone and in higher orders of organisation we characterise in detail. Effective promoter organizations are commonly found in the E. coli genome and are highly conserved between strains. This will allow studies of gene regulation at a previously unprecedented level of detail, where our framework can create testable hypothesis of promoter logic.

Transcription factors (TFs) are proteins that bind to specific sites on the DNA and regulate gene activity. Identifying where TF molecules bind and how much time they spend on their target sites is key to understanding transcriptional regulation. It is usually assumed that the free energy of binding of a TF to the DNA (the affinity of the site) is highly correlated to the amount of time the TF remains bound (the occupancy of the site). However, knowing the binding energy is not sufficient to infer actual binding site occupancy. This mismatch between the occupancy predicted by the affinity and the observed occupancy may be caused by various factors, such as TF abundance, competition between TFs or the arrangement of the sites on the DNA. We investigated the relationship between the affinity of a TF for a set of binding sites and their occupancy. In particular, we considered the case of the transcription factor lac repressor (lacI) in E.coli, and performed stochastic simulations of the TF dynamics on the DNA for various combinations of lacI abundance and competing TFs that contribute to macromolecular crowding. We also investigated the relationship of site occupancy and the information content of position weight matrices (PWMs) used to represent binding sites. Our results showed that for medium and high affinity sites, TF competition does not play a significant role for genomic occupancy except in cases when the abundance of the TF is significantly increased, or when the PWM displays relatively low information content. Nevertheless, for medium and low affinity sites, an increase in TF abundance (for both cognate and non-cognate molecules) leads to an increase in occupancy at several sites.

Transcription factor (TF) molecules translocate by facilitated diffusion (a combination of 3D diffusion around and 1D random walk on the DNA). Despite the attention this mechanism received in the last 40 years, only a few studies investigated the influence of the cellular environment on the facilitated diffusion mechanism and, in particular, the influence of `other' DNA binding proteins competing with the TF molecules for DNA space. Molecular crowding on the DNA is likely to influence the association rate of TFs to their target site and the steady state occupancy of those sites, but it is still not clear how it influences the search in a genome-wide context, when the model includes biologically relevant parameters (such as: TF abundance, TF affinity for DNA and TF dynamics on the DNA). We performed stochastic simulations of TFs performing the facilitated diffusion mechanism, and considered various abundances of cognate and non-cognate TFs. We show that, for both obstacles that move on the DNA and obstacles that are fixed on the DNA, changes in search time are not statistically significant in case of biologically relevant crowding levels on the DNA. In the case of non-cognate proteins that slide on the DNA, molecular crowding on the DNA always leads to statistically significant lower levels of occupancy, which may confer a general mechanism to control gene activity levels globally. When the `other' molecules are immobile on the DNA, we found a completely different behaviour, namely: the occupancy of the target site is always increased by higher molecular crowding on the DNA. Finally, we show that crowding on the DNA may increase transcriptional noise through increased variability of the occupancy time of the target sites.