Emmi Jokinen

TL;DR: In this article, a Gaussian process method was proposed to predict if TCRs recognize specified epitopes, which can utilize the amino acid sequences of the complementarity determining regions (CDRs) from TCRα and TCRβ chains and learn which CDRs are important in recognizing different epitopes.

TL;DR: A novel Gaussian process method to predict if TCRs recognize certain epitopes, which outperforms other state-of-the-art methods in epitope-specificity predictions is developed and is found in HBV-epitope specific T cells and their transcriptomic states in hepatocellular carcinoma patients.

TL;DR: In this paper, a Gaussian process method was proposed to predict if TCRs recognize certain epitopes, which can utilize CDR sequences from TCRα and TCRβ chains and learn which CDRs are important in recognizing different epitopes.

TL;DR: In this paper, a Gaussian Process (GP) based method for predicting protein's stability changes upon single and multiple mutations is proposed. But the accuracy of predictive models is ultimately constrained by the limited availability of experimental data.

TL;DR: Synthetic utility of DERA aldolase was improved by protein engineering approaches, and a novel machine learning model utilising Gaussian processes and feature learning was applied for the 3rd mutagenesis round to predict new beneficial mutant combinations.