PDGLin
Research Associate in the Pathogen Dynamics Group, Department of Genetics, University of Cambridge
University of CambridgeCambridge, UK
Pinned Repositories
AntigenicMapping
we develop a two-stage deep learning based framework to predict the antigenic distance between pairwise influenza viruses using their hemagglutinin (HA) protein sequences. In the first stage, for each type of influenza virus, we encode HA sequences in a high-dimensional continuous space by training a sequence processing model using BiLSTM with large-volume HA sequences retrieved from GISAID, the global influenza data initiative. In the second stage, for each influenza subtype/lineage, we train a ResNet to predict the antigenic distance between pairwise viruses using their HA sequence-based embedding distance.
CSD3
Submit R jobs using Cambridge HPC
kpp_drivers_dengue
COVID19_EffSerialInterval_NPI
CSD3
Submit R jobs using Cambridge HPC
DENVRisk
Code and data necessary to reproduce the analyses in "Linking antigenic diversity to dengue disease risk". https://www.researchsquare.com/article/rs-3214507/v1
nCoV2019
Location for summaries and analysis of data related to n-CoV 2019, first reported in Wuhan, China
StanPractice
PDGLin's Repositories
PDGLin/COVID19_EffSerialInterval_NPI
PDGLin/CSD3
Submit R jobs using Cambridge HPC
PDGLin/DENVRisk
Code and data necessary to reproduce the analyses in "Linking antigenic diversity to dengue disease risk". https://www.researchsquare.com/article/rs-3214507/v1
PDGLin/nCoV2019
Location for summaries and analysis of data related to n-CoV 2019, first reported in Wuhan, China
PDGLin/StanPractice