temileke Goto Github PK

ad_scrnaseq_guidline

guidelines for scRNAseq analysis

amyloidbeta_and_exocytosis

bi-be-cs-183-2023

Introduction to Computational Biology and Bioinformatics Course at Caltech, 2023

brian-material

This repository serves as a container for material around the Brian simulator, such as presentations and tutorials.

ca3bouton

Detailed spatial model of CA3 bouton

caiman

Computational toolbox for large scale Calcium Imaging Analysis, including movie handling, motion correction, source extraction, spike deconvolution and result visualization.

clei2block

Deep-neural protein translation

cosyne-tutorial-2022

Cosyne workshop tutorial 2022

course-content

Summer course content for Neuromatch Academy

deep-image-decoding-from-neural-activity

Explaining and predicting behavior from neural activity has been a longstanding goal in neuroscience. It is known that visual information is encoded within the hierarchical structure of the visual cortex, and plays an essential role in visual processing. However, the decoding of stimulus images from neural activity is still a challenging topic. Here we ask whether neural activity in the visual cortex of mice can be used to decode stimulus images, and whether specific visual cortex subregions recreate the images better than others. We hypothesize that neurons in the primary visual cortex (VISp) would best recreate these images. To investigate this, we employed a decoding approach outlined in previous literature. We obtained image visual features from a pre-trained deep residual neural network (ResNet), and created a linear mapping to corresponding neural activity (spike counts). This was then used to reconstruct the stimulus images through a generative adversarial network (GAN)-type layer. We observed that our model successfully decoded stimulus images from neural activity within a 70% accuracy. In addition, we found that VISp neurons achieve greater decoding quality relative to other subregions (80%). We conclude that our model can be used to accurately reconstruct stimulus images from neuronal spike counts, and that neuronal activity in the VISp encoded the majority of the information. Our findings may inspire a simple yet effective architecture for novel brain-computer-interface applications. Since our dataset contained a limited number of images and neuronal responses from one subject, generalization may be limited. We also have not examined whether combinations of subregions can recreate images better than single

deepimagereconstruction

Data and demo codes for Shen, Horikawa, Majima, and Kamitani (2019) Deep image reconstruction from human brain activity. PLoS Comput. Biol. http://dx.doi.org/10.1371/journal.pcbi.1006633.

demon-neurohack-challenge-1

er_and_exocytosis

gms7930

regulatory genomic sequencing

insilico

kinetic-neurotransmission

Python implementation of the kinetic model of neuromuscular transmission dynamics.

libra

machine-learning

mathtoolsforneuroscience

Materials for Mathematical Tools for Neuroscience course at Harvard (Neurobio 212)

matlab

Matlab codes for analysis of electrophysiology, imaging and simulation data

mlreview_notebooks

Jupyter notebooks for "A high-bias, low-variance introduction to Machine Learning for physicists"

nma-project

Trial-by-Trial Variability

nsci801-quantneuro

NSCI 801 (Queen's U) Quantitative Neuroscience course materials

project-rosalind-bioinformatics-stronghold

Python and T-SQL Rosalind Solutions

python_for_neuromatch

rbo

Python implementation of the rank-biased overlap list similarity measure.

rosalind

Solution to problems in https://rosalind.info/problems

scanpy

Single-cell analysis in Python. Scales to >1M cells.

scanpy-tutorials

Scanpy Tutorials.

scgen

Single cell perturbation prediction