paper reading list

Connectome

connectome dataset

MICrONS Project

technology of connectome

connection reconsturction

connectome analysis

FT(T;X)=∑c∈C(X)(nc&TnT)2FT(T;X) = \sum_{c \in C(X)} (\frac{n_{c \And T}}{n_T})^2

connectome and gene

review

computational methods

experimental test

data:

other methods to reconstruct connectome

human connectome

alt text|200pt

experiment support

cell-cell communication and interaction

protein interaction

experiment support

deep learning methods

result verification

alt text|200pt

data mapping between transcriptome and connectome

neural type

my blog on connectivity_to_type

p(y∣Q,f)∝e−12f′Mf+y′Qfp(\mathbf{y} \mid Q, \mathbf{f}) \propto e^{-\frac{1}{2} \mathbf{f}^{\prime} \mathbf{M f}+\mathbf{y}^{\prime} Q \mathbf{f}}

neural activity and connectivity

morphology

morphology data

NeuroMorpho.Org is a centrally curated inventorys of digitally reconstructed neurons and glia

maximum weighted independent set (MWIS)

Graph compression or sparsification

Maximum Dispersion

Maximum Diversity Problem (MDP)

connectome visualization

human connectome

gene data

gene data for Drosophila

analysis

logical gene for connectome

gene regulatory network

 Gene pairs  scCOPs ¬ scCOPs  bulkCOPs 3132993¬ bulkCOPs 3245232603\begin{array}{|c|c|c|} \hline \text { Gene pairs } & \text { scCOPs } & \neg \text { scCOPs } \\ \hline \text { bulkCOPs } & 313 & 2993 \\ \hline \neg \text { bulkCOPs } & 3245 & 232603 \\ \hline \end{array}

review paper

sota models

application

gene normalization

functional and structure conenctome

other connectome model:

Reinforcement learning (RL)

graph

https://paperswithcode.com/task/link-prediction

review

Graph Embedding

review

| Dataset | Nodes | Node Types | Edges | Edge Types | Target |
|---------|-------|------------|-------|------------|--------|
| Amazon  | 10,099| 1          | 148,659| 2          | product-product |
| LastFM  | 20,612| 3          | 141,521| 3          | user-artist |
| PubMed  | 63,109| 4          | 244,986| 10         | disease-disease |

disease gene

graph and connectome

BrainGNN: Interpretable Brain Graph Neural Network for fMRI Analysis https://www.sciencedirect.com/science/article/abs/pii/S1361841521002784

Graph Neural Networks for Brain Graph Learning: A Survey https://arxiv.org/abs/2406.02594 2024

dynamics system & Ripple

connectome inspired neural network
html connectome inspired neural network

-A model of a CA3 hippocampal pyramidal neuron incorporating voltage-clamp data on intrinsic conductances R. D. Traub,R. K. Wong,R. Miles, andH. Michelson, 1991

Soma compartments and neurite compartments

Spiking neural network

single neuron dynamics

U˙=−gLU+gD(Vw−U)+IUsom IUsom (t)=gE(t)(EE−U)+gI(t)(EI−U).\dot{U}=-g_{\mathrm{L}} U+g_{\mathrm{D}}\left(V_{\mathbf{w}}-U\right)+I_U^{\text {som }} \\ I_U^{\text {som }}(t)=g_{\mathrm{E}}(t)\left(E_{\mathrm{E}}-U\right)+g_{\mathrm{I}}(t)\left(E_{\mathrm{I}}-U\right) .

ddtui(x)=−ui(x)−ELτx+gxf(ui(d))+cxK(t−t^i(s))+Ii(x)+wi(x)Cxddtwi(x)=−wi(x)/τw(x)+aw(x)(ui(x)−EL)/τw(x)+bw(x)Si(x),\begin{gathered} \frac{d}{d t} u_i^{(x)}=-\frac{u_i^{(x)}-E_L}{\tau_x}+\frac{g_x f\left(u_i^{(d)}\right)+c_x K\left(t-\hat{t}_i^{(s)}\right)+I_i^{(x)}+w_i^{(x)}}{C_x} \\ \frac{d}{d t} w_i^{(x)}=-w_i^{(x)} / \tau_w^{(x)}+a_w^{(x)}\left(u_i^{(x)}-E_L\right) / \tau_w^{(x)}+b_w^{(x)} S_i^{(x)}, \end{gathered}

Models constrained only by BAC firing may fail to respond properly to perisomatic step current.|200pt

HH model

Hodgkin & Huxley 1952

LIF

GLIF model

synapse

We derive an effective point neuron model, which incorporates an additional synaptic integration current arising from the nonlinear interaction between synaptic currents across spatial dendrites.

We then derive a form of synaptic integration current within the point neuron framework to capture dendritic effects.

hLN

experiment support

multiple recent studies indicate that dendritic activity in vivo is much less compartmentalized than what was previously hypothesized (Xu et al., 2012, Beaulieu-Laroche et al., 2019, Francioni et al., 2019, Kerlin et al., 2019, Voigts and Harnett, 2019).

The first demonstrated that somatic voltage fluctuations under an in vivo-like synaptic regime are well described by a global dendritic nonlinearity that can capture up to 90% of explained variance (Ujfalussy et al., 2018).

Michalis Pagkalos, Spyridon Chavlis & Panayiota Poirazi, 2023, Nature Communications

tool

Tool for neuron dynamics simulation (like brian2, spikingjelly)

review

temporal inhibition
timing-dependent inhibition

modeling:

other

olfactory system

mean field

NeuroAI

Connectome & ANN

connectome inspired neural network
html connectome inspired neural network

geometry

Brain simulation / activity prediction

model

Network analysis

BOOK Fundamentals of Brain Network Analysis

BOOK Chapter 3 - Connectivity Matrices and Brain Graphs

Review Neural Networks With Motivation

synapse plasticity

STDP

nearest neighbor STDP (Izhikevich, 2003)

BTSP

neural data recording

whole brain imaging

drosophila data

other data

cell tracking

wormid

neural data analysis & modeling

reivew

high dimensional data

dimensional reduction

manifold reconstriction

activity analysis

head direction

encoding and decoding

deep learning method

fundation model

diffusion model/ visual Stimulus

behavior

oscilation

place cell and grid cell

vision

drosophila

others

ML

hippocamus

gene

cognition

attention

learning and memory

decision making

DS & OR

Quadratic assignment problem

similarity matrix

ML

NAS

CNN kernel

Clustering convolutional kernels

sparse decoding

UFLDL Tutorial

others

wiki

TODO