Biology subjects

Wen, H.

Publications and source records attributed to Wen, H..

10 recordsLinked to original sources

bioRxiv2018

Mechanisms of gene death in the Red Queen race revealed by the analysis of de novo microRNAs

The prevalence of de novo coding genes is controversial due to the length and coding constraints. Non-coding genes, especially small ones, are freer to evolve de novo by comparison. The best examples are microRNAs (miRNAs), a large class of regulatory molecules ~22 nt in length. Here, we study 6 de novo miRNAs in Drosophila which, like most new genes, are testis-specific. We ask how and why de novo genes die because gene death must be sufficiently frequent to balance the many new births. By knocking out each miRNA gene, we could analyze their contributions to each of the 9 components of male fitness (sperm production, length, competitiveness etc.). To our surprise, the knockout mutants often perform better in some components, and slightly worse in others, than the wildtype. When two of the younger miRNAs are assayed in long-term laboratory populations, their total fitness contributions are found to be essentially zero. These results collectively suggest that adaptive de novo genes die regularly, not due to the loss of functionality, but due to the canceling-out of positive and negative fitness effects, which may be characterized as \"quasi-neutrality\". Since de novo genes often emerge adaptively and become lost later, they reveal ongoing period-specific adaptations, reminiscent of the \"Red-Queen\" metaphor for long term evolution.

evolutionary biology↗

bioRxiv2018

Testing the Red Queen hypothesis on de novo new genes - Run or die in the evolution of new microRNAs

The Red Queen hypothesis depicts evolution as the continual struggle to adapt. According to this hypothesis, new genes, especially those originating from non-genic sequences (i.e., de novo genes), are eliminated unless they evolve continually in adaptation to a changing environment. Here, we analyze two Drosophila de novo miRNAs that are expressed in a testis-specific manner with very high rates of evolution in their DNA sequence. We knocked out these miRNAs in two sibling species and investigated their contributions to different fitness components. We observed that the fitness contributions of miR-975 in D. simulans seem positive, in contrast to its neutral contributions in D. melanogaster, while miR-983 appears to have negative contributions in both species, as the fitness of the knockout mutant increases. As predicted by the Red Queen hypothesis, the fitness difference of these de novo miRNAs indicates their different fates.

evolutionary biology↗

bioRxiv2018

Quasi-neutral molecular evolution -- When positive and negative selection cancel out

In the absence of both positive and negative selection, DNA sequences evolve at the neutral rate, R = 1. Due to the prevalence of negative selection, R[~]1 is rarely achieved in organismal evolution. However, when R [~] 1 is observed, it does not necessarily indicate neutral evolution because positive and negative selection could be equally strong but in opposite directions - hereby referred to as quasi-neutrality. We now show that somatic-cell evolution could be the paradigm of quasi-neutral evolution for these reasons: 1) Quasi-neutrality is much more likely in small populations (size N < 50) than in large ones; 2) Stem cell population sizes in single niches of normal tissues, from which tumors likely emerges, have small Ns (usually < 50); 3) the genome-wide evolutionary rate across tissue types is close to R = 1; 4) Relative to the average of R [~] 1, many genes evolve at a much higher or lower rate, thus hinting both positive and negative selection; 5) When N < 50, selection efficacy decreases rapidly as N decreases even when the selection intensity stays constant; 6) Notably, N is smaller in the small intestine (SmI) than in the colon (CO); hence, the [~] 70 fold higher rate of phenotypic evolution (observed as cancer risk) in the latter can be explained by the greater efficacy of selection, which then leads to the fixation of more advantageous mutations and fewer deleterious ones in the CO. Under quasineutrality, positive and negative selection can be measured in the same system as the two forces are simultaneously present or absent.

evolutionary biology↗

bioRxiv2018

Task-Evoked Functional Connectivity Does Not Explain Functional Connectivity Differences Between Rest and Task Conditions

During complex tasks, patterns of functional connectivity (FC) differ from those in the resting state. What accounts for such differences remains unclear. Brain activity during a task reflects an unknown mixture of spontaneous activity and task-evoked responses. The difference in FC between a task state and resting state may reflect not only task-evoked connectivity, but also changes in spontaneously emerging networks. Here, we characterized the difference in apparent functional connectivity between the resting state and when human subjects were watching a naturalistic movie. Such differences were marginally (3-15%) explained by the task-evoked networks directly involved in processing the movie content, but mostly attributable to changes in spontaneous networks driven by ongoing activity during the task. The execution of the task reduced the correlations in ongoing activity among different cortical networks, especially between the visual and non-visual sensory cortices. Our results suggest that the interaction between spontaneous and task-evoked activities is not mutually independent or linearly additive, and that engaging in a task may suppress ongoing activity.

neuroscience↗

bioRxiv2017

Variational Autoencoder: An Unsupervised Model for Modeling and Decoding fMRI Activity in Visual Cortex

Goal-driven convolutional neural networks (CNN) have been shown to be able to predict and decode cortical responses to natural images or videos. Here, we explored an alternative deep neural network, variational auto-encoder (VAE), as a computational model of the visual cortex. We trained a VAE with a five-layer encoder and a five-layer decoder to learn visual representations from a diverse set of unlabeled images. Inspired by the \"free-energy principle\" in neuroscience, we modeled the brains bottom-up and top-down pathways using the VAEs encoder and decoder, respectively. Following such conceptual relationships, we found that the VAE was able to predict cortical activities observed with functional magnetic resonance imaging (fMRI) from three human subjects watching natural videos. Compared to CNN, VAE resulted in relatively lower prediction accuracies, especially for higher-order ventral visual areas. On the other hand, fMRI responses could be decoded to estimate the VAEs latent variables, which in turn could reconstruct the visual input through the VAEs decoder. This decoding strategy was more advantageous than alternative decoding methods based on partial least square regression. This study supports the notion that the brain, at least in part, bears a generative model of the visual world.

neuroscience↗

bioRxiv2017

Deep Recurrent Neural Network Reveals a Hierarchy of Process Memory during Dynamic Natural Vision

The human visual cortex extracts both spatial and temporal visual features to support perception and guide behavior. Deep convolutional neural networks (CNNs) provide a computational framework to model cortical representation and organization for spatial visual processing, but unable to explain how the brain processes temporal information. To overcome this limitation, we extended a CNN by adding recurrent connections to different layers of the CNN to allow spatial representations to be remembered and accumulated over time. The extended model, or the recurrent neural network (RNN), embodied a hierarchical and distributed model of process memory as an integral part of visual processing. Unlike the CNN, the RNN learned spatiotemporal features from videos to enable action recognition. The RNN better predicted cortical responses to natural movie stimuli than the CNN, at all visual areas especially those along the dorsal stream. As a fully-observable model of visual processing, the RNN also revealed a cortical hierarchy of temporal receptive window, dynamics of process memory, and spatiotemporal representations. These results support the hypothesis of process memory, and demonstrate the potential of using the RNN for in-depth computational understanding of dynamic natural vision.

neuroscience↗

bioRxiv2017

Transferring and Generalizing Deep-Learning-based Neural Encoding Models across Subjects

Recent studies have shown the value of using deep learning models for mapping and characterizing how the brain represents and organizes information for natural vision. However, modeling the relationship between deep learning models and the brain (or encoding models), requires measuring cortical responses to large and diverse sets of natural visual stimuli from single subjects. This requirement limits prior studies to few subjects, making it difficult to generalize findings across subjects or for a population. In this study, we developed new methods to transfer and generalize encoding models across subjects. To train encoding models specific to a subject, the models trained for other subjects were used as the prior models and were refined efficiently using Bayesian inference with a limited amount of data from the specific subject. To train encoding models for a population, the models were progressively trained and updated with incremental data from different subjects. For the proof of principle, we applied these methods to functional magnetic resonance imaging (fMRI) data from three subjects watching tens of hours of naturalistic videos, while deep residual neural network driven by image recognition was used to model the visual cortical processing. Results demonstrate that the methods developed herein provide an efficient and effective strategy to establish subject-specific or population-wide predictive models of cortical representations of high-dimensional and hierarchical visual features.

neuroscience↗

bioRxiv2017

An improved experimental model of cystic hydatid disease in liver resembling natural infection route with stable growing dynamics and immune reaction

Cystic echinococcosis is an endemic parasitic infection in Xinjiang, China and is causing serious economic burdens and public health concerns. An experimental murine model in vivo for hepatic cystic echinococcosis was established in C57B/6 mice by injection with human protoscolices via the portal vein of three different concentrations. Mice were followed up 10 months by ultrasound, gross anatomy, pathological and immunological examinations. The protoscolice migration in portal vein, hydatid cyst growth, host immune reaction and hepatic histopathology were examed periodicly. The infection rate of the mice in the high, medium, and low concentration groups were 90%, 100%, and 63.6%, respectively. The protoscolices migrate in the portal vein with blood flow, settle in the liver and develop into orthotopic hepatic hydatid cysts, resembling the natural infection route and course. This study established an improved experimental model of low biohazard risk but stable growing dynamics and immune reaction. It is especially useful for new anti-parasite medication trials agains hydatid disease.\n\nSummary statementAn experimental murine model of cystic echinococcosis was set up. This orthotopic model resembles primary infection route and natural infectious course with low biohazard risk and high efficiency.

pathology↗

bioRxiv2017

Deep Residual Network Reveals a Nested Hierarchy of Distributed Cortical Representation for Visual Categorization

The brain represents visual objects with topographic cortical patterns. To address how distributed visual representations enable object categorization, we established predictive encoding models based on a deep residual neural network, and trained them to predict cortical responses to natural movies. Using this predictive model, we mapped human cortical representations to 64,000 visual objects from 80 categories with high throughput and accuracy. Such representations covered both the ventral and dorsal pathways, reflected multiple levels of object features, and preserved semantic relationships between categories. In the entire visual cortex, object representations were modularly organized into three categories: biological objects, non-biological objects, and background scenes. In a finer scale specific to each module, object representations revealed sub-modules for further categorization. These findings suggest that increasingly more specific category is represented by cortical patterns in progressively finer spatial scales. Such a nested hierarchy may be a fundamental principle for the brain to categorize visual objects with various levels of specificity, and can be explained and differentiated by object features at different levels.

neuroscience↗

bioRxiv2017

Musical Imagery Involves the Wernicke’s Area in Bilateral and Anti-Correlated Network Interactions in Musicians

Musical imagery is a human experience of imagining music without actually hearing it. The neural basis of such a mental ability is unclear, especially for musicians capable of accurate and vivid musical imagery due to their musical training. Here, we visualized an 8-min symphony as a silent movie, and used it as real-time cues for musicians to continuously imagine the music for multiple synchronized sessions during functional magnetic resonance imaging. The activations and networks evoked by musical imagery were compared with those when the subjects directly listened to the same music. The musical imagery and perception shared similar responses at bilateral secondary auditory areas and Wernickes area for encoding the musical feature. But the Wernickes area was involved in highly distinct network interactions during musical imagery vs. perception. The former involved positive correlations with a subset of the auditory network and the attention network, but negative correlations with the default mode network; the latter was confined to the intrinsic auditory network in the resting state. Our results highlight the important role of the Wernickes area in forming vivid musical imagery through bilateral and anti-correlated network interactions, challenging the conventional view of segregated and lateralized processing of music vs. language.

neuroscience↗