BZPEERPapers, people, evidence and research paths
Menu

Discover

SearchFind papers, people, topics, institutions and opportunities from one place.GraphSee how papers, authors, topics and evidence connect.

Research tools

HomeContinue from the papers, people and topics that matter now.CollectionsTurn saved papers into reading lists, evidence maps and shareable context.ResearchReconstruct the path that led you from search to paper to question.CollaborationMove from a paper, topic or expert request into a focused thread.PublishCreate full-text research articles with metadata, formulas and versions.

Network

ExpertsFind specialists whose work and reviews explain why they can help.CommunitiesJoin research groups organized around papers, topics and shared evidence.

Opportunities

ListingsBrowse roles, calls and collaborations connected to real research context.

Account

Log inReturn to your saved papers, graph views and active threads.Create accountStart saving papers, building a research trail and joining teams.
Log inCreate account

Source author record

Zhaokun Zhou

Zhaokun Zhou appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record
Biological PhysicsBiomoleculesComputer VisionNeural and Evolutionary ComputingMachine Learningphysics.optics

Catalog footprint

What is connected

6works
6topics
4close collaborators

Actions

Connect this record

Log in to claim
Open graphBrowse works

Research graph

See the researcher in context

Open full explorer

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

6 published item(s)

preprint2024arXiv

Parallel Spiking Neurons with High Efficiency and Ability to Learn Long-term Dependencies

Vanilla spiking neurons in Spiking Neural Networks (SNNs) use charge-fire-reset neuronal dynamics, which can only be simulated serially and can hardly learn long-time dependencies. We find that when removing reset, the neuronal dynamics can be reformulated in a non-iterative form and parallelized. By rewriting neuronal dynamics without reset to a general formulation, we propose the Parallel Spiking Neuron (PSN), which generates hidden states that are independent of their predecessors, resulting in parallelizable neuronal dynamics and extremely high simulation speed. The weights of inputs in the PSN are fully connected, which maximizes the utilization of temporal information. To avoid the use of future inputs for step-by-step inference, the weights of the PSN can be masked, resulting in the masked PSN. By sharing weights across time-steps based on the masked PSN, the sliding PSN is proposed to handle sequences of varying lengths. We evaluate the PSN family on simulation speed and temporal/static data classification, and the results show the overwhelming advantage of the PSN family in efficiency and accuracy. To the best of our knowledge, this is the first study about parallelizing spiking neurons and can be a cornerstone for the spiking deep learning research. Our codes are available at \url{https://github.com/fangwei123456/Parallel-Spiking-Neuron}.

0 citations0 reviewsNo code linkedOpen access
preprint2024arXiv

Spikformer V2: Join the High Accuracy Club on ImageNet with an SNN Ticket

Spiking Neural Networks (SNNs), known for their biologically plausible architecture, face the challenge of limited performance. The self-attention mechanism, which is the cornerstone of the high-performance Transformer and also a biologically inspired structure, is absent in existing SNNs. To this end, we explore the potential of leveraging both self-attention capability and biological properties of SNNs, and propose a novel Spiking Self-Attention (SSA) and Spiking Transformer (Spikformer). The SSA mechanism eliminates the need for softmax and captures the sparse visual feature employing spike-based Query, Key, and Value. This sparse computation without multiplication makes SSA efficient and energy-saving. Further, we develop a Spiking Convolutional Stem (SCS) with supplementary convolutional layers to enhance the architecture of Spikformer. The Spikformer enhanced with the SCS is referred to as Spikformer V2. To train larger and deeper Spikformer V2, we introduce a pioneering exploration of Self-Supervised Learning (SSL) within the SNN. Specifically, we pre-train Spikformer V2 with masking and reconstruction style inspired by the mainstream self-supervised Transformer, and then finetune the Spikformer V2 on the image classification on ImageNet. Extensive experiments show that Spikformer V2 outperforms other previous surrogate training and ANN2SNN methods. An 8-layer Spikformer V2 achieves an accuracy of 80.38% using 4 time steps, and after SSL, a 172M 16-layer Spikformer V2 reaches an accuracy of 81.10% with just 1 time step. To the best of our knowledge, this is the first time that the SNN achieves 80+% accuracy on ImageNet. The code will be available at Spikformer V2.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

Force spectroscopy in studying infection

Biophysical force spectroscopy tools - for example optical tweezers, magnetic tweezers, atomic force microscopy, - have been used to study elastic, mechanical, conformational and dynamic properties of single biological specimens from single proteins to whole cells to reveal information not accessible by ensemble average methods such as X-ray crystallography, mass spectroscopy, gel electrophoresis and so on. Here we review the application of these tools on a range of infection-related questions from antibody-inhibited protein processivity to virus-cell adhesion. In each case we focus on how the instrumental design tailored to the biological system in question translates into the functionality suitable for that particular study. The unique insights that force spectroscopy has gained to complement knowledge learned through population averaging techniques in interrogating biomolecular details prove to be instrumental in therapeutic innovations such as those in structure-based drug design.

0 citations0 reviewsNo code linkedOpen access
preprint2015arXiv

Developing a new biophysical tool to combine magneto-optical tweezers with super-resolution fluorescence microscopy

We present a novel experimental setup in which magnetic and optical tweezers are combined for torque and force transduction onto single filamentous molecules in a transverse configuration to allow simultaneous mechanical measurement and manipulation. Previously we have developed a super-resolution imaging module which in conjunction with advanced imaging techniques such as Blinking assisted Localisation Microscopy (BaLM) achieves localisation precision of single fluorescent dye molecules bound to DNA of ~30 nm along the contour of the molecule; our work here describes developments in producing a system which combines tweezing and super-resolution fluorescence imaging. The instrument also features an acousto-optic deflector that temporally divides the laser beam to form multiple traps for high throughput statistics collection. Our motivation for developing the new tool is to enable direct observation of detailed molecular topological transformation and protein binding event localisation in a stretching/twisting mechanical assay that previously could hitherto only be deduced indirectly from the end-to-end length variation of DNA. Our approach is simple and robust enough for reproduction in the lab without the requirement of precise hardware engineering, yet is capable of unveiling the elastic and dynamic properties of filamentous molecules that have been hidden using traditional tools.

0 citations0 reviewsNo code linkedOpen access
preprint2015arXiv

Probing DNA interactions with proteins using a single-molecule toolbox: inside the cell, in a test tube, and in a computer

DNA-interacting proteins have roles multiple processes, many operating as molecular machines which undergo dynamic metastable transitions to bring about their biological function. To fully understand this molecular heterogeneity, DNA and the proteins that bind to it must ideally be interrogated at a single molecule level in their native in vivo environments, in a time-resolved manner fast to sample the molecular transitions across the free energy landscape. Progress has been made over the past decade in utilising cutting-edge tools of the physical sciences to address challenging biological questions concerning the function and modes of action of several different proteins which bind to DNA. These physiologically relevant assays are technically challenging, but can be complemented by powerful and often more tractable in vitro experiments which confer advantages of the chemical environment with enhanced detection single-to-noise of molecular signatures and transition events. Here, we discuss a range of techniques we have developed to monitor DNA-protein interactions in vivo, in vitro and in silico. These include bespoke single-molecule fluorescence microscopy techniques to elucidate the architecture and dynamics of the bacterial replisome and the structural maintenance of bacterial chromosomes, as well as new computational tools to extract single-molecule molecular signatures from live cells to monitor stoichiometry, spatial localization and mobility in living cells. We also discuss recent developments from our lab made in vitro, complementing these in vivo studies, which combine optical and magnetic tweezers to manipulate and image single molecules of DNA, with and without bound protein, in a new superresolution fluorescence microscope.

0 citations0 reviewsNo code linkedOpen access
preprint2015arXiv

Superresolution imaging of single DNA molecules using stochastic photoblinking of minor groove and intercalating dyes

As proof-of-principle for generating superresolution structural information from DNA we applied a method of localization microscopy utilizing photoblinking comparing intercalating dye YOYO-1 against minor groove binding dye SYTO-13, using a bespoke multicolor single-molecule fluorescence microscope. We used a full-length ~49 kbp λ DNA construct possessing oligo inserts at either terminus allowing conjugation of digoxigenin and biotin at opposite ends for tethering to a glass coverslip surface and paramagnetic microsphere respectively. We observed stochastic DNA-bound dye photoactivity consistent with dye photoblinking as opposed to binding/unbinding events, evidenced through both discrete simulations and continuum kinetics analysis. We analyzed dye photoblinking images of immobilized DNA molecules using superresolution reconstruction software from two existing packages, rainSTORM and QuickPALM, and compared the results against our own novel home-written software called ADEMS code. ADEMS code generated lateral localization precision values of 30-40 nm and 60-70 nm for YOYO-1 and SYTO-13 respectively at video-rate sampling, similar to rainSTORM, running more slowly than rainSTORM and QuickPALM algorithms but having a complementary capability over both in generating automated centroid distribution and cluster analyses. Our imaging system allows us to observe dynamic topological changes to single molecules of DNA in real-time, such as rapid molecular snapping events. This will facilitate visualization of fluorescently-labeled DNA molecules conjugated to a magnetic bead in future experiments involving newly developed magneto-optical tweezers combined with superresolution microscopy.

0 citations0 reviewsNo code linkedOpen access