A5033557144- Neural dynamics and brain function
- Neuroscience and Neural Engineering
- Advanced biosensing and bioanalysis techniques
- Advanced Memory and Neural Computing
- Advanced Fluorescence Microscopy Techniques
- CRISPR and Genetic Engineering
- Molecular Biology Techniques and Applications
- Cell Image Analysis Techniques
- Single-cell and spatial transcriptomics
- RNA and protein synthesis mechanisms
- Neural Networks and Applications
- Force Microscopy Techniques and Applications
- EEG and Brain-Computer Interfaces
- Photoreceptor and optogenetics research
- RNA Research and Splicing
- Advanced Biosensing Techniques and Applications
- Biomedical and Engineering Education
- Reinforcement Learning in Robotics
- Genetics, Bioinformatics, and Biomedical Research
- Lipid Membrane Structure and Behavior
- Microfluidic and Bio-sensing Technologies
- Microfluidic and Capillary Electrophoresis Applications
- Genomics and Phylogenetic Studies
- Quantum Information and Cryptography
- Advanced Electron Microscopy Techniques and Applications
Massachusetts Institute of Technology
2014-2025
McGovern Institute for Brain Research
2024-2025
Convergent Engineering (United States)
2024
Human Media
2016-2023
Engineering Arts (United States)
2021-2022
Federation of American Scientists
2021-2022
Google (United States)
2018-2020
DeepMind (United Kingdom)
2019
Harvard University
2012-2018
Harvard Bioscience (United States)
2012-2014
DNA nanotechnology exploits the programmable specificity afforded by base-pairing to produce self-assembling macromolecular objects of custom shape. For building megadalton-scale nanostructures, a long 'scaffold' strand can be employed template assembly hundreds oligonucleotide 'staple' strands into planar antiparallel array cross-linked helices. We recently adapted this 'scaffolded origami' method producing 3D shapes formed as pleated layers double helices constrained honeycomb lattice....
Transcripts Visualized in Situ Despite advances, current methods for single-cell sequencing are unable to resolve transcript location within the cell, so Lee et al. (p. 1360 , published online 27 February) developed a method of fluorescent situ RNA (FISSEQ) that works vivo show messenger localization cells. The amplifies complementary DNA targets by rolling circle amplification, and then cross-linking locks amplicons produce ample, highly localized templates three-dimensional sequencing....
Neuroscience has focused on the detailed implementation of computation, studying neural codes, dynamics and circuits. In machine learning, however, artificial networks tend to eschew precisely designed or circuits in favor brute force optimization a cost function, often using simple relatively uniform initial architectures. Two recent developments have emerged within learning that create an opportunity connect these seemingly divergent perspectives. First, structured architectures are used,...
Identifying transcript location in cells where specific RNAs occur within a cell or tissue has been limited by technology and imaging capabilities. Expansion microscopy allowed for better visualization of small structures expanding the tissues with polymer- hydrogel-based system. Alon et al. combined expansion long-read situ RNA sequencing, resulting more precise transcripts. This method, termed “ExSeq” was used to detect RNAs, both new transcripts those previously demonstrated localize...
Simultaneously measuring the activities of all neurons in a mammalian brain at millisecond resolution is challenge beyond limits existing techniques neuroscience. Entirely new approaches may be required, motivating an analysis fundamental physical constraints on problem. We outline principles governing activity mapping using optical, electrical,magnetic resonance, and molecular modalities neural recording. Focusing mouse brain, we analyze scalability each method, concentrating limitations...
Shrinking problems in 3D printing Although a range of materials can now be fabricated using additive manufacturing techniques, these usually involve assembly series stacked layers, which restricts three-dimensional (3D) geometry. Oran et al. developed method to print materials, including metals and semiconductors, inside gel scaffold (see the Perspective by Long Williams). When hydrogels were dehydrated, they shrunk 10-fold, pushed feature sizes down nanoscale. Science , this issue p. 1281 ;...
Abstract Lipid membranes are key to the nanoscale compartmentalization of biological systems, but fluorescent visualization them in intact tissues, with precision, is challenging do high labeling density. Here, we report ultrastructural membrane expansion microscopy (umExM), which combines an innovative label and optimized protocol, support dense tissues for visualization. We validate signal-to-background ratio, uniformity continuity, umExM brain slices, supports imaging proteins at a...
Based on a survey of the literature, we attempt to answer Frequently Asked Questions issues cortical uniformity vs. non-uniformity, neural mechanisms symbolic variable binding, and other highlighted in (Marcus, Marblestone Dean. "The Atoms Neural Computation". Science. 31 October 2014. Vol 346. Issue 6209).
Abstract Lipids are fundamental building blocks of cells and their organelles, yet nanoscale resolution imaging lipids has been largely limited to electron microscopy techniques. We introduce validate a chemical tag that enables lipid membranes be imaged optically at via lipid-optimized form expansion microscopy, which we call membrane (mExM). mExM, novel post-expansion antibody labeling protocol, protein-lipid relationships in organelles such as mitochondria, the endoplasmic reticulum,...
Abstract Lipid membranes are key to the nanoscale compartmentalization of biological systems, but fluorescent visualization them in intact tissues, with precision, is challenging do high labeling density. Here, we report ultrastructural membrane expansion microscopy (umExM), which combines a novel label and optimized protocol, support dense tissues for visualization. We validated signal-to-background ratio, uniformity continuity, umExM brain slices, supported imaging proteins at resolution...
High-throughput recording of signals embedded within inaccessible micro-environments is a technological challenge. The ideal device would be nanoscale machine capable quantitatively transducing wide range variables into molecular medium suitable for long-term storage and facile readout in the form digital data. We have recently proposed such device, which cation concentrations modulate misincorporation rate DNA polymerase (DNAP) on known template, allowing sequences to encode information...
Neuroscience has long been an essential driver of progress in artificial intelligence (AI). We propose that to accelerate AI, we must invest fundamental research NeuroAI. A core component this is the embodied Turing test, which challenges AI animal models interact with sensorimotor world at skill levels akin their living counterparts. The test shifts focus from those capabilities like game playing and language are especially well-developed or uniquely human capabilities, inherited over 500...
We present the theory of a Josephson parametric amplifier employing two-pump sources. Our calculations are based on input-output theory, and can easily be generalized to any coupled system involving interactions. analyze operation device, taking into account feedback introduced by reaction signal noise pump power, in this framework, compute response functions interest---signal idler gains, internal gain amplifier, self-oscillation amplitude. To for back action between pump, we adopt...
Neuroscience has focused on the detailed implementation of computation, studying neural codes, dynamics and circuits. In machine learning, however, artificial networks tend to eschew precisely designed or circuits in favor brute force optimization a cost function, often using simple relatively uniform initial architectures. Two recent developments have emerged within learning that create an opportunity connect these seemingly divergent perspectives. First, structured architectures are used,...
Advances in sensing technology raise the possibility of creating neural interfaces that can more effectively restore or repair function and reveal fundamental properties information processing. To realize potential these bioelectronic devices, it is necessary to understand capabilities emerging technologies identify best strategies translate into products therapies will improve lives patients with neurological other disorders. Here we discuss for brain activity, anticipated challenges...
Abstract: Methods for highly multiplexed RNA imaging are limited in spatial resolution, and thus their ability to localize transcripts nanoscale subcellular compartments. We adapt expansion microscopy, which physically expands biological specimens, long-read untargeted targeted situ sequencing. applied sequencing (ExSeq) mouse brain, yielding readout of thousands genes, including splice variants novel transcripts. Targeted ExSeq yielded nanoscale-resolution maps RNAs throughout dendrites...
A molecular device that records time-varying signals would enable new approaches in neuroscience. We have recently proposed such a device, termed "molecular ticker tape", which an engineered DNA polymerase (DNAP) writes into the form of nucleotide misincorporation patterns. Here, we define theoretical framework quantifying expected capabilities tapes as function experimental parameters. present decoding algorithm for estimating time-dependent input signals, and DNAP kinetic parameters,...
Abstract De novo synthesis of long double‐stranded DNA constructs has a myriad applications in biology and biological engineering. However, its widespread adoption been hindered by high costs. Cost can be significantly reduced using oligonucleotides synthesized on high‐density chips. most methods for off‐chip gene have failed to scale due the error rates, low yields, chemical complexity chip‐synthesized oligonucleotides. We recently demonstrated that some commercial chip manufacturers...
DNA polymerase fidelity is affected by both intrinsic properties and environmental conditions. Current strategies for measuring error rate in vitro are constrained low subtype sensitivity, poor scalability, lack of flexibility types sequence contexts that can be tested. We have developed the Magnification via Nucleotide Imbalance Fidelity (MagNIFi) assay, a scalable next-generation sequencing assay uses biased deoxynucleotide pool to quantitatively shift rates into range where errors...
We propose and theoretically study an approach to massively parallel single molecule peptide sequencing, based on measurement of the kinetics probe binding (Havranek, et al., 2013) N-termini immobilized peptides. Unlike previous proposals, this method is robust both weak non-specific probe-target affinities, which we demonstrate by applying a range randomized affinity matrices consisting relatively low-quality binders. This suggests novel principle for proteomic whereby highly non-optimized...