Atmospheric processes

Research group

Research theme

Sub department

About

Postprocessing East African rainfall forecasts using a generative machine learning model

(2024)

Authors:

Bobby Antonio, Andrew McRae, David MacLeod, Fenwick Cooper, John Marsham, Laurence Aitchison, Tim Palmer, Peter Watson
More details from the publisher

A Machine Learning Approach for Predicting Essentiality of Metabolic Genes

In: Braman, J.C. (eds) Synthetic Biology. Methods in Molecular Biology, vol 2760 (2024)

Authors:

Lilli J Freischem & Diego A Oyarzún

Abstract:

The identification of essential genes is a key challenge in systems and synthetic biology, particularly for engineering metabolic pathways that convert feedstocks into valuable products. Assessment of gene essentiality at a genome scale requires large and costly growth assays of knockout strains. Here we describe a strategy to predict the essentiality of metabolic genes using binary classification algorithms. The approach combines elements from genome-scale metabolic models, directed graphs, and machine learning into a predictive model that can be trained on small knockout data. We demonstrate the efficacy of this approach using the most complete metabolic model of Escherichia coli and various machine learning algorithms for binary classification.
More details from the publisher

Predictable decadal forcing of the North Atlantic jet speed by sub-polar North Atlantic sea surface temperatures

Weather and Climate Dynamics Copernicus Publications 4:4 (2023) 853-874

Authors:

Kristian Strommen, Tim Woollings, Paolo Davini, Paolo Ruggieri, Isla R Simpson
More details from the publisher
More details

Using probabilistic machine learning to better model temporal patterns in parameterizations: a case study with the Lorenz 96 model

Geoscientific Model Development Copernicus Publications 16:15 (2023) 4501-4519

Authors:

Raghul Parthipan, Hannah M Christensen, J Scott Hosking, Damon J Wischik
More details from the publisher

On the relationship between reliability diagrams and the ‘signal-to-noise paradox’

Geophysical Research Letters American Geophysical Union 50:14 (2023) e2023GL103710

Authors:

Kristian Strommen, Molly MacRae, Hannah Christensen

Abstract:

The ‘signal-to-noise paradox’ for seasonal forecasts of the winter NAO is often described as an ‘underconfident’ forecast and measured using the ratio-of-predictable components metric (RPC). However, comparison of RPC with other measures of forecast confidence, such as spread-error ratios, can give conflicting impressions, challenging this informal description. We show, using a linear statistical model, that the ‘paradox’ is equivalent to a situation where the reliability diagram of any percentile forecast has a slope exceeding 1. The relationship with spread-error ratios is shown to be far less direct. We furthermore compute reliability diagrams of winter NAO forecasts using seasonal hindcasts from the European Centre for Medium-range Weather Forecasts and the UK Meteoro logical Office. While these broadly exhibit slopes exceeding 1, there is evidence of asymmetry between upper and lower terciles, indicating a potential violation of linearity/Gaussianity. The limitations and benefits of reliability diagrams as a diagnostic tool are discussed.
More details from the publisher
Details from ORA
More details