@article{11638, abstract = {Statistical inference is central to many scientific endeavors, yet how it works remains unresolved. Answering this requires a quantitative understanding of the intrinsic interplay between statistical models, inference methods, and the structure in the data. To this end, we characterize the efficacy of direct coupling analysis (DCA)—a highly successful method for analyzing amino acid sequence data—in inferring pairwise interactions from samples of ferromagnetic Ising models on random graphs. Our approach allows for physically motivated exploration of qualitatively distinct data regimes separated by phase transitions. We show that inference quality depends strongly on the nature of data-generating distributions: optimal accuracy occurs at an intermediate temperature where the detrimental effects from macroscopic order and thermal noise are minimal. Importantly our results indicate that DCA does not always outperform its local-statistics-based predecessors; while DCA excels at low temperatures, it becomes inferior to simple correlation thresholding at virtually all temperatures when data are limited. Our findings offer insights into the regime in which DCA operates so successfully, and more broadly, how inference interacts with the structure in the data.}, author = {Ngampruetikorn, Vudtiwat and Sachdeva, Vedant and Torrence, Johanna and Humplik, Jan and Schwab, David J. and Palmer, Stephanie E.}, issn = {2643-1564}, journal = {Physical Review Research}, number = {2}, publisher = {American Physical Society}, title = {{Inferring couplings in networks across order-disorder phase transitions}}, doi = {10.1103/PhysRevResearch.4.023240}, volume = {4}, year = {2022}, } @article{12156, abstract = {Models of transcriptional regulation that assume equilibrium binding of transcription factors have been less successful at predicting gene expression from sequence in eukaryotes than in bacteria. This could be due to the non-equilibrium nature of eukaryotic regulation. Unfortunately, the space of possible non-equilibrium mechanisms is vast and predominantly uninteresting. The key question is therefore how this space can be navigated efficiently, to focus on mechanisms and models that are biologically relevant. In this review, we advocate for the normative role of theory—theory that prescribes rather than just describes—in providing such a focus. Theory should expand its remit beyond inferring mechanistic models from data, towards identifying non-equilibrium gene regulatory schemes that may have been evolutionarily selected, despite their energy consumption, because they are precise, reliable, fast, or otherwise outperform regulation at equilibrium. We illustrate our reasoning by toy examples for which we provide simulation code.}, author = {Zoller, Benjamin and Gregor, Thomas and Tkačik, Gašper}, issn = {2452-3100}, journal = {Current Opinion in Systems Biology}, keywords = {Applied Mathematics, Computer Science Applications, Drug Discovery, General Biochemistry, Genetics and Molecular Biology, Modeling and Simulation}, number = {9}, publisher = {Elsevier}, title = {{Eukaryotic gene regulation at equilibrium, or non?}}, doi = {10.1016/j.coisb.2022.100435}, volume = {31}, year = {2022}, } @article{10530, abstract = {Cell dispersion from a confined area is fundamental in a number of biological processes, including cancer metastasis. To date, a quantitative understanding of the interplay of single cell motility, cell proliferation, and intercellular contacts remains elusive. In particular, the role of E- and N-Cadherin junctions, central components of intercellular contacts, is still controversial. Combining theoretical modeling with in vitro observations, we investigate the collective spreading behavior of colonies of human cancer cells (T24). The spreading of these colonies is driven by stochastic single-cell migration with frequent transient cell-cell contacts. We find that inhibition of E- and N-Cadherin junctions decreases colony spreading and average spreading velocities, without affecting the strength of correlations in spreading velocities of neighboring cells. Based on a biophysical simulation model for cell migration, we show that the behavioral changes upon disruption of these junctions can be explained by reduced repulsive excluded volume interactions between cells. This suggests that in cancer cell migration, cadherin-based intercellular contacts sharpen cell boundaries leading to repulsive rather than cohesive interactions between cells, thereby promoting efficient cell spreading during collective migration. }, author = {Zisis, Themistoklis and Brückner, David and Brandstätter, Tom and Siow, Wei Xiong and d’Alessandro, Joseph and Vollmar, Angelika M. and Broedersz, Chase P. and Zahler, Stefan}, issn = {0006-3495}, journal = {Biophysical Journal}, keywords = {Biophysics}, number = {1}, pages = {P44--60}, publisher = {Elsevier}, title = {{Disentangling cadherin-mediated cell-cell interactions in collective cancer cell migration}}, doi = {10.1016/j.bpj.2021.12.006}, volume = {121}, year = {2022}, } @article{10736, abstract = {Predicting function from sequence is a central problem of biology. Currently, this is possible only locally in a narrow mutational neighborhood around a wildtype sequence rather than globally from any sequence. Using random mutant libraries, we developed a biophysical model that accounts for multiple features of σ70 binding bacterial promoters to predict constitutive gene expression levels from any sequence. We experimentally and theoretically estimated that 10–20% of random sequences lead to expression and ~80% of non-expressing sequences are one mutation away from a functional promoter. The potential for generating expression from random sequences is so pervasive that selection acts against σ70-RNA polymerase binding sites even within inter-genic, promoter-containing regions. This pervasiveness of σ70-binding sites implies that emergence of promoters is not the limiting step in gene regulatory evolution. Ultimately, the inclusion of novel features of promoter function into a mechanistic model enabled not only more accurate predictions of gene expression levels, but also identified that promoters evolve more rapidly than previously thought.}, author = {Lagator, Mato and Sarikas, Srdjan and Steinrueck, Magdalena and Toledo-Aparicio, David and Bollback, Jonathan P and Guet, Calin C and Tkačik, Gašper}, issn = {2050-084X}, journal = {eLife}, publisher = {eLife Sciences Publications}, title = {{Predicting bacterial promoter function and evolution from random sequences}}, doi = {10.7554/eLife.64543}, volume = {11}, year = {2022}, } @article{12332, abstract = {Activity of sensory neurons is driven not only by external stimuli but also by feedback signals from higher brain areas. Attention is one particularly important internal signal whose presumed role is to modulate sensory representations such that they only encode information currently relevant to the organism at minimal cost. This hypothesis has, however, not yet been expressed in a normative computational framework. Here, by building on normative principles of probabilistic inference and efficient coding, we developed a model of dynamic population coding in the visual cortex. By continuously adapting the sensory code to changing demands of the perceptual observer, an attention-like modulation emerges. This modulation can dramatically reduce the amount of neural activity without deteriorating the accuracy of task-specific inferences. Our results suggest that a range of seemingly disparate cortical phenomena such as intrinsic gain modulation, attention-related tuning modulation, and response variability could be manifestations of the same underlying principles, which combine efficient sensory coding with optimal probabilistic inference in dynamic environments.}, author = {Mlynarski, Wiktor F and Tkačik, Gašper}, issn = {1545-7885}, journal = {PLoS Biology}, number = {12}, pages = {e3001889}, publisher = {Public Library of Science}, title = {{Efficient coding theory of dynamic attentional modulation}}, doi = {10.1371/journal.pbio.3001889}, volume = {20}, year = {2022}, }