Results 81 to 90 of about 26,557 (311)

Reconstructing enzyme evolution by protein engineering

FEBS Letters, EarlyView.
Natural enzyme evolution can be retraced by protein engineering methods such as directed evolution, rational design, and ancestral sequence reconstruction. These approaches reveal how enzymes emerged from ligand‐binding scaffolds, developed varying substrate preferences, formed oligomeric complexes, adapted to environmental changes, and evolved novel ...
Lukas Drexler, Markus R. Busch, Reinhard Sterner   +2 more
wiley   +1 more source

Graph Communal Contrastive Learning

Proceedings of the ACM Web Conference 2022, 2022
Graph representation learning is crucial for many real-world applications (e.g. social relation analysis). A fundamental problem for graph representation learning is how to effectively learn representations without human labeling, which is usually costly and time-consuming. Graph contrastive learning (GCL) addresses this problem by pulling the positive
Bolian Li, Baoyu Jing, Hanghang Tong
openaire   +2 more sources

Interpreting the effects of DNA polymerase variants at the structural level

Molecular Oncology, EarlyView.
Using MAVISp and molecular dynamics simulations, we analyzed over 60 000 missense variants in POLE and POLD1 from ClinVar, COSMIC, cBioPortal, and saturation mutagenesis. Identified mechanistic indicators, including stability, binding, and long‐range, enable structural interpretation, providing ACMG‐like evidence for possible reclassification of VUS ...
Matteo Arnaudi, Karolina Krzesińska, Ludovica Beltrame, Pablo Sánchez‐Izquierdo Besora, Matteo Tiberti, Mef Nilbert, Anna Rohlin, Elena Papaleo   +7 more
wiley   +1 more source

Steel surface defect detection based on self-supervised contrastive representation learning with matching metric [PDF]

, 2023
Defect detection is crucial in the quality control of industrial applications. Existing supervised methods are heavily reliant on the large amounts of labeled data.
Hussain, Amir, Amir Hussain, Mandar Gogate, Dashtipour, Kia, Xuejin Hu, Fengling Jiang, Jiang, Fengling, Gogate, Mandar, Jing Yang, Yang, Jing, Kia Dashtipour, Hu, Xuejin   +11 more
core   +1 more source

Node classification in complex networks based on multi-view debiased contrastive learning

Complex & Intelligent Systems
In complex networks, contrastive learning has emerged as a crucial technique for acquiring discriminative representations from graph data. Maximizing the similarity among relevant sample pairs while minimizing that among irrelevant pairs is pivotal in ...
Zhe Li, Lei Zhou, Yandong Hou, Min Ji, Zhuanzheng Hang, Bolun Chen   +5 more
doaj   +1 more source

Developmental programmes drive cellular plasticity, disease progression and therapy resistance in lung adenocarcinoma

Molecular Oncology, EarlyView.
This study shows that lung adenocarcinomas exploit developmental branching morphogenesis to acquire a therapy resistant basal‐like tumour cell state. This process was found to be regulated by combined TP53 loss‐of‐function and type‐I interferon signalling, identifying a novel axis for biomarker and therapeutic target discovery.
Kamila J Bienkowska, Stephany Gallardo, Nur S. Zainal, Leena Arora, Matthew Ellis, Maria‐Antoinette Lopez, Judith Austine, Sai Pittla, Serena J Chee, Aiman Alzetani, Emily C Shaw, Christian H Ottensmeier, Gareth J Thomas, Christopher J Hanley   +13 more
wiley   +1 more source

Learning Representations with Contrastive Self-Supervised Learning for Histopathology Applications [Elektronisk resurs] [PDF]

, 2022
Unsupervised learning has made substantial progress over the last few years, especially by means of contrastive self-supervised learning. The dominating dataset for benchmarking self-supervised learning has been ImageNet, for which recent methods are ...
Stacke, Karin,, Eilertsen, Gabriel,, Lundström, Claes,, Unger, Jonas,   +3 more
core  

Contrastive Self-Supervised Learning for Sensor-Based Human Activity Recognition: A Review

IEEE Access
Deep learning models have achieved significant success in human activity recognition, particularly in assisted living and telemonitoring. However, training these models requires substantial amounts of labeled training data, which is time-consuming and ...
Hui Chen, Charles Gouin-Vallerand, Kevin Bouchard, Sebastien Gaboury, Melanie Couture, Nathalie Bier, Sylvain Giroux   +6 more
doaj   +1 more source

Graph Contrastive Learning for Materials

CoRR, 2022
Recent work has shown the potential of graph neural networks to efficiently predict material properties, enabling high-throughput screening of materials. Training these models, however, often requires large quantities of labelled data, obtained via costly methods such as ab initio calculations or experimental evaluation.
Teddy Koker, Keegan Quigley, Will Spaeth, Nathan C. Frey, Lin Li   +4 more
openaire   +2 more sources

Liquid biopsy‐based diagnostic evaluation of hypermethylated CpG sites for ovarian cancer diagnosis

Molecular Oncology, EarlyView.
This schematic outlines the workflow from biomarker identification to duplex MethyLight assay validation for epithelial ovarian cancer diagnosis using cfDNA‐based liquid biopsy. Initial screening of hypermethylated CpG candidates (cg02957270, cg10061138 cg00480298, COL2A1) was performed in tissue using ARMS‐PCR, COBRA, qPCR and image analysis. Selected
Deepa Bisht, Sameer Gupta, Amrita Chaurasia, Manisha Sachan   +3 more
wiley   +1 more source