Journal: Frontiers in Computational Neuroscience

Article Title: Contextual Modulation is Related to Efficiency in a Spiking Network Model of Visual Cortex

doi: 10.3389/fncom.2015.00155

Figure Lengend Snippet: The stimuli, predicted and simulated neural activation patterns. (A) The selected area of an example natural image illustrates the projection of the image to the Gabor filter and ASF calculations of the left visual field. Small dots in the pink square show the cortical density of cells in the visual field. The density of cells in each eccentricity follows the human cortical magnification factor. (B) The contrast energy image from the same natural image as in (A) , built as the output of a series of Gabor filters. These contrast images are the input for estimating the target area summation patters. (C) The area summation target activation pattern, corresponding to the same natural image as in (A) . The color code of the activation pattern borders illustrates the projection of the image from the visual field to the model cortex, with the color code as in (A) . (D) Simulated output pattern for the image in (A) . (E) The normalized strength as a function of the spatial frequency contents for the natural images in Low and Mixed frequencies groups. Error bars represent the standard error of mean. (F) The output of the retina filter to the natural image in (A) , which served as an input to the spiking neural network simulations.

Article Snippet: We simulated the neural responses of V1 in a biomimetic recurrent spiking network model, to study the relation between the non-linear area summation properties of primate V1 neurons and efficiency of the neural activity at the population level.

Techniques: Activation Assay

Journal: Frontiers in Computational Neuroscience

Article Title: Contextual Modulation is Related to Efficiency in a Spiking Network Model of Visual Cortex

doi: 10.3389/fncom.2015.00155

Figure Lengend Snippet: Simulation result compared to natural ASF . (A) Examples of the spiking network response associated with high (solid blue curve) and low (solid red curve) efficiency measures to an increasing stimulus patch size, for a representative model neuron located at 14° eccentricity. Response is averaged across 15 neighboring cells. The high-efficiency response is similar to ASF target at the same eccentricity (dashed curve) but with smaller summation field size. (B) Normalized mean distance to ASF (DAS), number of spikes, entropy per spike, and sparseness across all 20 natural images as a function of the number of V1 lateral excitatory-inhibitory and V1-extrastriate feedforward-feedback connections.

Article Snippet: We simulated the neural responses of V1 in a biomimetic recurrent spiking network model, to study the relation between the non-linear area summation properties of primate V1 neurons and efficiency of the neural activity at the population level.

Techniques:

Journal: BMC Genomics

Article Title: Illuminating the dark side of the human transcriptome with long read transcript sequencing

doi: 10.1186/s12864-020-07123-7

Figure Lengend Snippet: Long transcript assembly benchmarking. Sensitivity and precision of guided and unguided long read transcriptome assembly methods. a Gene loci level for simulated PacBio reads. b Gene loci level for simulated Nanopore reads. c Transcript level for simulated PacBio reads. d Transcript level for simulated Nanopore reads. e Transcript level for PacBio Sequel II Iso-Seq SIRV reads. f Example of erroneous transcript prediction by guided Talon where supporting reads are from another transcript model. These supporting reads are from 5′ degraded RNA resulting in the confusion

Article Snippet: To address the issues with simulated datasets, we also used reads from the Lexogen SIRV spike-in from the PacBio UHRR Sequel II Iso-Seq dataset.

Techniques: