A deep learning framework for quantitative analysis of actin microridges

Microridges are evolutionarily conserved actin-rich protrusions present on the apical surface of squamous epithelial cells. In zebrafish epidermal cells, microridges form self-evolving patterns due to the underlying actomyosin network dynamics. However, their morphological and dynamic characteristics have remained poorly understood owing to a lack of computational methods. We achieved ~95% pixel-level accuracy with a deep learning microridge segmentation strategy enabling quantitative insights into their bio-physical-mechanical characteristics. From the segmented images, we estimated an effective microridge persistence length of ~6.1 μm. We discovered the presence of mechanical fluctuations and found relatively greater stresses stored within patterns of yolk than flank, indicating distinct regulation of their actomyosin networks. Furthermore, spontaneous formations and positional fluctuations of actin clusters within microridges were associated with pattern rearrangements over short length/time-scales. Our framework allows large-scale spatiotemporal analysis of microridges during epithelial development and probing of their responses to chemical and genetic perturbations to unravel the underlying patterning mechanisms.

Spatial correlations in a finite-range Kuramoto model

We study spatial correlations between oscillator phases in the steady state of a Kuramoto model, in which phase oscillators that are randomly distributed in space interact with constant strength but within a limited range. Such a model could be relevant, for example, in the synchronization of gene expression oscillations in cells, where only oscillations of neighboring cells are coupled through cell-cell contacts. We analytically infer spatial phase-phase correlation functions from the known steady-state distribution of oscillators for the case of homogenous frequencies and show that these can contain information about the range and strength of interactions, provided the noise in the system can be estimated. We suggest a method for the latter, and also explore when correlations appear to be ergodic in this model, which would enable an experimental measurement of correlation functions to utilize temporal averages. Simulations show that our techniques also give qualitative results for the model with heterogenous frequencies. We illustrate our results by comparison with experimental data on genetic oscillations in the segmentation clock of zebrafish embryos.

Segmentation clock dynamics is strongly synchronized in the forming somite

During vertebrate somitogenesis an inherent segmentation clock coordinates the spatiotemporal signaling to generate segmented structures that pattern the body axis. Using our experimental and quantitative approach, we study the cell movements and the genetic oscillations of her1 expression level at single-cell resolution simultaneously and scale up to the entire pre-somitic mesoderm (PSM) tissue. From the experimentally determined phases of PSM cellular oscillators, we deduced an in vivo frequency profile gradient along the anterior-posterior PSM axis and inferred precise mathematical relations between spatial cell-level period and tissue-level somitogenesis period. We also confirmed a gradient in the relative velocities of cellular oscillators along the axis. The phase order parameter within an ensemble of oscillators revealed the degree of synchronization in the tailbud and the posterior PSM being only partial, whereas synchronization can be almost complete in the presumptive somite region but with temporal oscillations. Collectively, the degree of synchronization itself, possibly regulated by cell movement and the synchronized temporal phase of the transiently expressed clock protein Her1, can be an additional control mechanism for making precise somite boundaries.

A framework for quantification and physical modeling of cell mixing applied to oscillator synchronization in vertebrate somitogenesis

In development and disease, cells move as they exchange signals. One example is found in vertebrate development, during which the timing of segment formation is set by a ‘segmentation clock’, in which oscillating gene expression is synchronized across a population of cells by Delta-Notch signaling. Delta-Notch signaling requires local cell-cell contact, but in the zebrafish embryonic tailbud, oscillating cells move rapidly, exchanging neighbors. Previous theoretical studies proposed that this relative movement or cell mixing might alter signaling and thereby enhance synchronization. However, it remains unclear whether the mixing timescale in the tissue is in the right range for this effect, because a framework to reliably measure the mixing timescale and compare it with signaling timescale is lacking. Here, we develop such a framework using a quantitative description of cell mixing without the need for an external reference frame and constructing a physical model of cell movement based on the data. Numerical simulations show that mixing with experimentally observed statistics enhances synchronization of coupled phase oscillators, suggesting that mixing in the tailbud is fast enough to affect the coherence of rhythmic gene expression. Our approach will find general application in analyzing the relative movements of communicating cells during development and disease.

Summary: We develop a framework to quantify and model cell mixing independent of a choice of reference frames, and apply this to study oscillator synchronization in the zebrafish segmentation clock.

In silico detection of binding mode of J-superfamily conotoxin pl14a with Kv1.6 channel

A novel conotoxin pl14a containing 25 amino acid residues with an amidated C-terminus from vermivorous cone snail, Conus planorbis belongs to J-conotoxin superfamily and this is the first conotoxin, which inhibits both nicotinic acetylcholine receptor subtypes and Kv1.6 channel. We have attempted through bioinformatics approaches to elucidate the extent of specificity of pl14a towards Kv1 channel subtypes (Kv1.1-Kv1.6). Our work provides rationale for the relatively high specificity and binding mode of pl14a to Kv1.6 channel. The pl14a peptide contains two types of structural elements, namely the putative dyad (Lys18 and Tyr19) and basic residue ring constituted of arginine residues. We have carried out in silico docking studies so as to assess the contribution of one or combination of both structural elements of pl14a in blocking of Kv1.6 channel. For this purpose, we have built by homology modelling, the theoretical 3D structure of Kv1.6 channel based on the available crystal structure of mammalian shaker Kv1.2 channel. Docking studies suggest that positively charged residues ring may be involved in the blocking mechanism of Kv1.6 channel. The models suggest that the peptide interacts with negatively charged extracellular loops and pore-mouth of the potassium channel and blocks the channel by covering the pore as a lid, akin to previously proposed blocking mechanism of kappaM-conotoxin RIIIK from Conus radiatus to Tsha1 potassium channel. The newly detected pharmacophore for pl14a interacting with Kv1.6 channel provides a pointer to experimental work to validate the observations made here. Based on differences in the number and distribution of the positively-charged residues in other conopeptides from the J-superfamily, we hypothesize different selectivity profiles against subtypes of the potassium channels for these conopeptides.

Pseudo amino acid composition and multi-class support vector machines approach for conotoxin superfamily classification

Conotoxins are disulfide rich small peptides that target a broad spectrum of ion-channels and neuronal receptors. They offer promising avenues in the treatment of chronic pain, epilepsy and cardiovascular diseases. Assignment of newly sequenced mature conotoxins into appropriate superfamilies using a computational approach could provide valuable preliminary information on the biological and pharmacological functions of the toxins. However, creation of protein sequence patterns for the reliable identification and classification of new conotoxin sequences may not be effective due to the hypervariability of mature toxins. With the aim of formulating an in silico approach for the classification of conotoxins into superfamilies, we have incorporated the concept of pseudo-amino acid composition to represent a peptide in a mathematical framework that includes the sequence-order effect along with conventional amino acid composition. The polarity index attribute, which encodes information such as residue surface buriability, polarity, and hydropathy, was used to store the sequence-order effect. Several methods like BLAST, ISort (Intimate Sorting) predictor, least Hamming distance algorithm, least Euclidean distance algorithm and multi-class support vector machines (SVMs), were explored for superfamily identification. The SVMs outperform other methods providing an overall accuracy of 88.1% for all correct predictions with generalized squared correlation of 0.75 using jackknife cross-validation test for A, M, O and T superfamilies and a negative set consisting of short cysteine rich sequences from different eukaryotes having diverse functions. The computed sensitivity and specificity for the superfamilies were found to be in the range of 84.0-94.1% and 80.0-95.5%, respectively, attesting to the efficacy of multi-class SVMs for the successful in silico classification of the conotoxins into their superfamilies.

I-conotoxin superfamily revisited

The I-conotoxin superfamily (I-Ctx) is known to have four disulfide bonds with the cysteine arrangement C-C-CC-CC-C-C, and the members inhibit or modify ion channels of nerve cells. Recently, Olivera and co-workers (FEBS J. 2005; 272: 4178-4188) have suggested that the previously described I-Ctx should now be divided into two different gene superfamilies, namely, I1 and I2, in view of their having two different types of signal peptides and exhibiting distinct functions. We have revisited the 28 entries presently grouped as I-Ctx in UniProt Swiss-Prot knowledgebase, and on the basis of in silico analysis have divided them into I1 and I2 superfamilies. The sequence analysis has provided a framework for in silico annotation enabling us to carry out computer-based functional characterization of the UniProtKB/TrEMBL entry Q59AA4 from Conus miles and to predict it as a member of the I2 superfamily. Furthermore, we have predicted the mature toxin of this entry and have proposed that it may be an inhibitor of voltage-gated potassium channels.