Systematic Perturbation of an Artificial Neural Network: A Step Towards Quantifying Causal Contributions in The Brain

Lesion inference analysis is a fundamental approach for characterizing the causal contributions of neural elements to brain function. Historically, it has helped to localize specialized functions in the brain after brain damage, and it has gained new prominence through the arrival of modern optogenetic perturbation techniques that allow probing the functional contributions of neural circuit elements at unprecedented levels of detail. While inferences drawn from brain lesions are conceptually powerful, they face methodological difficulties due to the brain's complexity. Particularly, they are challenged to disentangle the functional contributions of individual neural elements because many elements may contribute to a particular function, and these elements may be interacting anatomically as well as functionally. Therefore, studies of real-world data, as in clinical lesion studies, are not suitable for establishing the reliability of lesion approaches due to an unknown, potentially complex ground truth. Instead, ground truth studies of well-characterized artificial systems are required. Here, we systematically and exhaustively lesioned a small Artificial Neural Network (ANN) playing a classic arcade game. We determined the functional contributions of all nodes and links, contrasting results from single-element perturbations and perturbing multiple elements simultaneously. Moreover, we computed pairwise causal functional interactions between the network elements, and looked deeper into the system's inner workings, proposing a mechanistic explanation for the effects of lesions. We found that not every perturbation necessarily reveals causation, as lesioning elements, one at a time, produced biased results. By contrast, multi-site lesion analysis captured crucial details that were missed by single-site lesions. We conclude that even small and seemingly simple ANNs show surprising complexity that needs to be understood for deriving a causal picture of the system. In the context of rapidly evolving multivariate brain-mapping approaches and inference methods, we advocate using in-silico experiments and ground-truth models to verify fundamental assumptions, technical limitations, and the scope of possible interpretations of these methods.

One-loop correction to the Casimir energy in Lifshitz-like theory

In this paper, Radiative Correction (RC) to the Casimir energy was computed for the self-interacting massive/massless Lifshitz-like scalar field, confined between a pair of plates with Dirichlet and Mixed boundary conditions in 3 + 1 dimensions. Moreover, using the results obtained for the Dirichlet Casimir energy, the RC to the Casimir energy for Periodic and Neumann boundary conditions were also draw outed. To renormalize the bare parameters of the Lagrangian, a systematic perturbation expansion was used in which the counterterms were automatically obtained in a position-dependent manner. In our view, the position dependency of the counterterm was allowed, since it reflected the effects of the boundary condition imposed or the background space in the problem. All the answers obtained for the Casimir energy were consistent with well-known physical expects. In a language of graphs, the Casimir energy for the massive Lifshitz-like scalar field confined with four boundary conditions (Dirichlet, Neumann, Mixed, and Periodic) was also compared to each other, and as a concluding remark, the sign and magnitude of their values were discussed.

Endocytic vacuole formation by WASH-mediated endocytosis

Endocytosis of extracellular or plasma membrane material is a fundamental process. A variety of endocytic pathways exist, several of which are barely understood in terms of mechanistic execution and biological function. Importantly, some mechanisms have been identified and characterized by following virus internalization into cells. This includes a novel endocytic pathway exploited by human papillomavirus type 16 (HPV16). However, its cellular role and mechanism of endocytic vacuole formation remain unclear. Here, HPV16 was used as a tool to examine the mechanistic execution of vesicle formation by combining systematic perturbation of cellular processes with electron and video microscopy. Our results indicate cargo uptake by uncoated, inward-budding pits facilitated by the membrane bending retromer protein SNX2. Actin polymerization-driven vesicle scission is promoted by WASH, an actin regulator typically not found at the plasma membrane. Uncovering a novel role of WASH in endocytosis, we propose to term the new pathway WASH-mediated endocytosis (WASH-ME).

Phonon redshift and Hubble friction in an expanding BEC

We revisit the theoretical analysis of an expanding ring-shaped Bose-Einstein condensate. Starting from the action and integrating over dimensions orthogonal to the phonon’s direction of travel, we derive an effective one-dimensional wave equation for azimuthally-travelling phonons. This wave equation shows that expansion redshifts the phonon frequency at a rate determined by the effective azimuthal sound speed, and damps the amplitude of the phonons at a rate given by \dot{\mathcal{V}}/{\mathcal{V}}𝒱̇/𝒱, where \mathcal{V}𝒱 is the volume of the background condensate. This behavior is analogous to the redshifting and ``Hubble friction’’ for quantum fields in the expanding universe and, given the scalings with radius determined by the shape of the ring potential, is consistent with recent experimental and theoretical results. The action-based dimensional reduction methods used here should be applicable in a variety of settings, and are well suited for systematic perturbation expansions.

Periodically Forced Nonlinear Oscillators With Hysteretic Damping

We perform a detailed study of the dynamics of a nonlinear, one-dimensional oscillator driven by a periodic force under hysteretic damping, whose linear version was originally proposed and analyzed by Bishop (1955, “The Treatment of Damping Forces in Vibration Theory,” Aeronaut. J., 59(539), pp. 738–742). We first add a small quadratic stiffness term in the constitutive equation and construct the periodic solution of the problem by a systematic perturbation method, neglecting transient terms as t→∞. We then repeat the analysis replacing the quadratic by a cubic term, which does not allow the solutions to escape to infinity. In both cases, we examine the dependence of the amplitude of the periodic solution on the different parameters of the model and discuss the differences with the linear model. We point out certain undesirable features of the solutions, which have also been alluded to in the literature for the linear Bishop's model, but persist in the nonlinear case as well. Finally, we discuss an alternative hysteretic damping oscillator model first proposed by Reid (1956, “Free Vibration and Hysteretic Damping,” Aeronaut. J., 60(544), pp. 283–283), which appears to be free from these difficulties and exhibits remarkably rich dynamical properties when extended in the nonlinear regime.

CLUE: a bioinformatic and wet-lab pipeline for multiplexed cloning of custom sgRNA libraries

The systematic perturbation of genomes using CRISPR/Cas9 deciphers gene function at an unprecedented rate, depth and ease. Commercially available sgRNA libraries typically contain tens of thousands of pre-defined constructs, resulting in a complexity challenging to handle. In contrast, custom sgRNA libraries comprise gene sets of self-defined content and size, facilitating experiments under complex conditions such as in vivo systems. To streamline and upscale cloning of custom libraries, we present CLUE, a bioinformatic and wet-lab pipeline for the multiplexed generation of pooled sgRNA libraries. CLUE starts from lists of genes or pasted sequences provided by the user and designs a single synthetic oligonucleotide pool containing various libraries. At the core of the approach, a barcoding strategy for unique primer binding sites allows amplifying different user-defined libraries from one single oligonucleotide pool. We prove the approach to be straightforward, versatile and specific, yielding uniform sgRNA distributions in all resulting libraries, virtually devoid of cross-contaminations. For in silico library multiplexing and design, we established an easy-to-use online platform at www.crispr-clue.de. All in all, CLUE represents a resource-saving approach to produce numerous high quality custom sgRNA libraries in parallel, which will foster their broad use across molecular biosciences.

CLUE: a bioinformatic and wet-lab pipeline for multiplexed cloning of custom sgRNA libraries

The systematic perturbation of genomes using CRISPR/Cas9 deciphers gene function at an unprecedented rate, depth and ease. Commercially available sgRNA libraries typically contain tens of thousands of pre-defined constructs, resulting in a complexity challenging to handle. In contrast, custom sgRNA libraries comprise gene sets of self-defined content and size, facilitating experiments under complex conditions such as in vivo systems. To streamline and upscale cloning of custom libraries, we present CLUE, a bioinformatic and wet-lab pipeline for the multiplexed generation of pooled sgRNA libraries. CLUE starts from lists of genes or pasted sequences provided by the user and designs a single synthetic oligonucleotide pool containing various libraries. At the core of the approach, a barcoding strategy for unique primer binding sites allows amplifying different distinct libraries from one single oligonucleotide pool. We prove the approach to be straightforward, versatile and specific, yielding uniform sgRNA distributions in all resulting libraries, virtually devoid of cross-contaminations. For in silico library multiplexing and design, we established an easy-to-use online platform at www.crispr-clue.de. All in all, CLUE represents a resource-saving approach to produce numerous high quality custom sgRNA libraries in parallel, which will foster their broad use across molecular biosciences.

Systematic perturbation of retroviral LTRs reveals widespread long-range effects on human gene regulation

Recent work suggests extensive adaptation of transposable elements (TEs) for host gene regulation. However, high numbers of integrations typical of TEs, coupled with sequence divergence within families, have made systematic interrogation of the regulatory contributions of TEs challenging. Here, we employ CARGO, our recent method for CRISPR gRNA multiplexing, to facilitate targeting of LTR5HS, an ape-specific class of HERVK (HML-2) LTRs that is active during early development and present in ~700 copies throughout the human genome. We combine CARGO with CRISPR activation or interference to, respectively, induce or silence LTR5HS en masse, and demonstrate that this system robustly targets the vast majority of LTR5HS insertions. Remarkably, activation/silencing of LTR5HS is associated with reciprocal up- and down-regulation of hundreds of human genes. These effects require the presence of retroviral sequences, but occur over long genomic distances, consistent with a pervasive function of LTR5HS elements as early embryonic enhancers in apes.