Detection of native and mirror protein structures based on Ramachandran plot analysis by interpretable machine learning models

AbstractIn this contribution the discrimination between native and mirror models of proteins according to their chirality is tackled based on the structural protein information. This information is contained in the Ramachandran plots of the protein models. We provide an approach to classify those plots by means of an interpretable machine learning classifier - the Generalized Matrix Learning Vector Quantizer. Applying this tool, we are able to distinguish with high accuracy between mirror and native structures just evaluating the Ramachandran plots. The classifier model provides additional information regarding the importance of regions, e.g. α-helices and β-strands, to discriminate the structures precisely. This importance weighting differs for several considered protein classes.

Download Full-text

Advanced Interpretable Machine Learning Methods for Clinical NGS Big Data of Complex Hereditary Diseases

10.3389/978-2-88966-274-6 ◽

2020 ◽

Keyword(s):

Machine Learning ◽

Big Data ◽

Hereditary Diseases ◽

Learning Methods ◽

Machine Learning Methods ◽

Interpretable Machine Learning

Download Full-text

Interpretable Machine Learning: Shapley Values (Seminar Slides)

SSRN Electronic Journal ◽

10.2139/ssrn.3637020 ◽

2020 ◽

Author(s):

Marcos López de Prado

Keyword(s):

Machine Learning ◽

Interpretable Machine Learning ◽

Shapley Values

Download Full-text

Investable and Interpretable Machine Learning for Equities

SSRN Electronic Journal ◽

10.2139/ssrn.3747203 ◽

2020 ◽

Author(s):

Yimou Li ◽

Zachary Simon ◽

David Turkington

Keyword(s):

Machine Learning ◽

Interpretable Machine Learning

Download Full-text

Prediction of Streptococcus uberis clinical mastitis treatment success in dairy herds by means of mass spectrometry and machine-learning

Scientific Reports ◽

10.1038/s41598-021-87300-0 ◽

2021 ◽

Vol 11 (1) ◽

Author(s):

Alexandre Maciel-Guerra ◽

Necati Esener ◽

Katharina Giebel ◽

Daniel Lea ◽

Martin J. Green ◽

...

Keyword(s):

Machine Learning ◽

Mass Spectrometry ◽

Treatment Success ◽

Clinical Mastitis ◽

Supervised Machine Learning ◽

Structural Protein ◽

Cohen’S Kappa ◽

Maldi Tof ◽

Streptococcus Uberis ◽

Cohen's Kappa

AbstractStreptococcus uberis is one of the leading pathogens causing mastitis worldwide. Identification of S. uberis strains that fail to respond to treatment with antibiotics is essential for better decision making and treatment selection. We demonstrate that the combination of supervised machine learning and matrix-assisted laser desorption ionization/time of flight (MALDI-TOF) mass spectrometry can discriminate strains of S. uberis causing clinical mastitis that are likely to be responsive or unresponsive to treatment. Diagnostics prediction systems trained on 90 individuals from 26 different farms achieved up to 86.2% and 71.5% in terms of accuracy and Cohen’s kappa. The performance was further increased by adding metadata (parity, somatic cell count of previous lactation and count of positive mastitis cases) to encoded MALDI-TOF spectra, which increased accuracy and Cohen’s kappa to 92.2% and 84.1% respectively. A computational framework integrating protein–protein networks and structural protein information to the machine learning results unveiled the molecular determinants underlying the responsive and unresponsive phenotypes.

Download Full-text

Techniques for interpretable machine learning

Communications of the ACM ◽

10.1145/3359786 ◽

2019 ◽

Vol 63 (1) ◽

pp. 68-77 ◽

Cited By ~ 47

Author(s):

Mengnan Du ◽

Ninghao Liu ◽

Xia Hu

Keyword(s):

Machine Learning ◽

Interpretable Machine Learning

Download Full-text

Functional prediction of environmental variables using metabolic networks

Scientific Reports ◽

10.1038/s41598-021-91486-8 ◽

2021 ◽

Vol 11 (1) ◽

Author(s):

Adèle Weber Zendrera ◽

Nataliya Sokolovska ◽

Hédi A. Soula

Keyword(s):

Machine Learning ◽

Growth Temperature ◽

Environmental Variables ◽

Metabolic Networks ◽

Machine Learning Techniques ◽

Underlying Structure ◽

Glutathione Biosynthesis ◽

Additional Information ◽

Cold Environments ◽

Novel Approach

AbstractIn this manuscript, we propose a novel approach to assess relationships between environment and metabolic networks. We used a comprehensive dataset of more than 5000 prokaryotic species from which we derived the metabolic networks. We compute the scope from the reconstructed graphs, which is the set of all metabolites and reactions that can potentially be synthesized when provided with external metabolites. We show using machine learning techniques that the scope is an excellent predictor of taxonomic and environmental variables, namely growth temperature, oxygen tolerance, and habitat. In the literature, metabolites and pathways are rarely used to discriminate species. We make use of the scope underlying structure—metabolites and pathways—to construct the predictive models, giving additional information on the important metabolic pathways needed to discriminate the species, which is often absent in other metabolic network properties. For example, in the particular case of growth temperature, glutathione biosynthesis pathways are specific to species growing in cold environments, whereas tungsten metabolism is specific to species in warm environments, as was hinted in current literature. From a machine learning perspective, the scope is able to reduce the dimension of our data, and can thus be considered as an interpretable graph embedding.

Download Full-text