Mycorrhiza: Genotype Assignment usingPhylogenetic Networks
Authors:
Jeremy Georges-Filteau,
Richard C. Hamelin,
Mathieu Blanchette
Abstract:
Motivation The genotype assignment problem consists of predicting, from the genotype of an individual, which of a known set of populations it originated from. The problem arises in a variety of contexts, including wildlife forensics, invasive species detection and biodiversity monitoring. Existing approaches perform well under ideal conditions but are sensitive to a variety of common violations of…
▽ More
Motivation The genotype assignment problem consists of predicting, from the genotype of an individual, which of a known set of populations it originated from. The problem arises in a variety of contexts, including wildlife forensics, invasive species detection and biodiversity monitoring. Existing approaches perform well under ideal conditions but are sensitive to a variety of common violations of the assumptions they rely on. Results In this article, we introduce Mycorrhiza, a machine learning approach for the genotype assignment problem. Our algorithm makes use of phylogenetic networks to engineer features that encode the evolutionary relationships among samples. Those features are then used as input to a Random Forests classifier. The classification accuracy was assessed on multiple published empirical SNP, microsatellite or consensus sequence datasets with wide ranges of size, geographical distribution and population structure and on simulated datasets. It compared favorably against widely used assessment tests or mixture analysis methods such as STRUCTURE and Admixture, and against another machine-learning based approach using principal component analysis for dimensionality reduction. Mycorrhiza yields particularly significant gains on datasets with a large average fixation index (FST) or deviation from the Hardy-Weinberg equilibrium. Moreover, the phylogenetic network approach estimates mixture proportions with good accuracy.
△ Less
Submitted 13 October, 2020;
originally announced October 2020.
Ultra-fast perpendicular Spin Orbit Torque MRAM
Authors:
Murat Cubukcu,
Olivier Boulle,
Nikolaï Mikuszeit,
Claire Hamelin,
Thomas Brächer,
Nathalie Lamard,
Marie-Claire Cyrille,
Liliana Buda-Prejbeanu,
Kevin Garello,
Ioan Mihai Miron,
O. Klein,
G. de Loubens,
V. V. Naletov,
Juergen Langer,
Berthold Ocker,
Pietro Gambardella,
Gilles Gaudin
Abstract:
We demonstrate ultra-fast (down to 400 ps) bipolar magnetization switching of a three-terminal perpendicular Ta/FeCoB/MgO/FeCoB magnetic tunnel junction. The critical current density rises significantly as the current pulse shortens below 10 ns, which translates into a minimum in the write energy in the ns range. Our results show that SOT-MRAM allows fast and low power write operations, which rend…
▽ More
We demonstrate ultra-fast (down to 400 ps) bipolar magnetization switching of a three-terminal perpendicular Ta/FeCoB/MgO/FeCoB magnetic tunnel junction. The critical current density rises significantly as the current pulse shortens below 10 ns, which translates into a minimum in the write energy in the ns range. Our results show that SOT-MRAM allows fast and low power write operations, which renders it promising for non-volatile cache memory applications.
△ Less
Submitted 5 March, 2018; v1 submitted 8 September, 2015;
originally announced September 2015.
