Peter C. Doerschuk’s research while affiliated with Cornell University and other places

... In mRNA vaccine design, GCNs encode secondary and tertiary structures of mRNA as graph networks, where nodes represent nucleotides, and edges depict structural interactions, such as hydrogen bonds or stacking interactions. This representation enables the identification of key features, like stem-loops or pseudoknots, that contribute to stability and translation efficiency, facilitating the optimization of mRNA constructs for robust antigen expression (Dorsey et al., 2024). ...

Reference:

Computational biology and artificial intelligence in mRNA vaccine design for cancer immunotherapy

... Evaluating geometric disparities between simulated and experimental structures through computer-vision-based methodologies also requires computational efficiency and real-time capabilities. These algorithms encompass feature extraction, pattern recognition, and symmetric difference analysis [XDC24]. FPGAs, with inherent parallelism and low latency, could facilitate defect detection and expedite feedback during design and manufacturing. ...

Reference:

Enabling Heterogeneous Computing for Software Developers

... Hui et al. (2023) designed a neural net to predict quality and resistance of inkjet-printed silver lines in relation to various printing process parameters. Ivy et al. (2023) developed a ML-model to optically predict resistance of inkjet-printed resistors based on their geometry and texture features, using a high-resolution color scanner. Gafurov et al. (2022) utilized deep-learning algorithms for the optical quality assessment and parameter identification of screen-printed lines. ...

Reference:

Image-based identification of optical quality and functional properties in inkjet-printed electronics using machine learning

... Using the BotFactory SV2 thermal Inkjet PCB printer, another machine can simultaneously print conductive and DIs. In [141], laser-induced graphene sensors and their wiring and packaging were fabricated following the printing process shown in Fig. 9. To begin fabrication, each device's four sets of three alignment windows are printed onto a 125-µm-thick Kapton board. ...

Reference:

A Review of Multimaterial Additively Manufactured Electronics and 4-D Printing/Origami Shape-Memory Devices: Design, Fabrication, and Implementation

... 2. Apply the algorithm proposed in [89] to find an equivalent real representation, if the irrep provided by GAP is complex. ...

Reference:

Bootstrap for finite N lattice Yang-Mills theory

... In particular, the functional role of the regions could be affected by changes in brain structure. Indeed, recent work using fMRI has shown in humans the presence of a functional asymmetry between brain regions in terms of afferent and efferent information transfer 33 . Other computer modelling work has also shown a relationship between network topology and information directionality, in particular, by identifying certain brain regions (or nodes) as targets and sources of information 34 . ...

Reference:

Linking structural and functional changes during aging using multilayer brain network analysis

... Past research has explored the directed connectivity patterns in the human brain using different methods (Bajaj et al., 2016;Chén et al., 2019;Duggento et al., 2018;Lund et al., 2020;Schwab et al., 2018;Xu et al., 2020), although no one has studied how heritability influences the "information flow" from a functional network to another one. ...

Reference:

Heritability of human "directed" functional connectome

... Cage-like breakable organosilica NPs were synthesized and characterized following a previously published protocol [46][47][48][49]. Cetyltrimethylammonium bromide (CTAB) was used as the template of the cage-like NPs and tetraethoxysilane (TEOS) and bis3-(triethoxysilyl) propyl]disulphide BTDS were employed for synthetizing the disulfide breakable organosilica framework. ...

Reference:

Melanoma extracellular vesicles membrane coated nanoparticles as targeted delivery carriers for tumor and lungs

... Our goal is to further merge the ideas that biological particles obey symmetry and that different instances of the particle are heterogeneous due to different vibrational states (continuous heterogeneity). In the continuous heterogeneity situation, it becomes more natural to impose the symmetry on the 1st-and 2nd-order statistics of \rho (x) for the particle (symmetric statistics) [65,64], rather than on the \rho (x) itself. In this more realistic assumption, since only the statistics have symmetry, the individual particles may be nonsymmetric, and therefore, the invariant basis becomes no longer sufficient. ...

Reference:

Computation of Real-Valued Basis Functions which Transform as Irreducible Representations of the Polyhedral Groups

... where L βvae is the encoder-decoder loss of disentangled VAE, also called beta-VAE [27]. beta-VAE is a modified version of vanilla VAE and can provide better disentanglement of distribution in the latent space [28]. The beta-VAE training objective is given by, ...

Reference:

U-TELL: Unsupervised Task Expert Lifelong Learning