Inertial Sensor Location for Ground Reaction Force and Gait Event Detection Using Reservoir Computing in Gait.

Inertial measurement units (IMUs) have shown promising outcomes for estimating gait event detection (GED) and ground reaction force (GRF). This study aims to determine the best sensor location for GED and GRF prediction in gait using data from IMUs for healthy and medial knee osteoarthritis (MKOA) individuals. In this study, 27 healthy and 18 MKOA individuals participated. Participants walked at different speeds on an instrumented treadmill. Five synchronized IMUs (Physilog®, 200 Hz) were placed on the lower limb (top of the shoe, heel, above medial malleolus, middle and front of tibia, and on medial of shank close to knee joint). To predict GRF and GED, an artificial neural network known as reservoir computing was trained using combinations of acceleration signals retrieved from each IMU. For GRF prediction, the best sensor location was top of the shoe for 72.2% and 41.7% of individuals in the healthy and MKOA populations, respectively, based on the minimum value of the mean absolute error (MAE). For GED, the minimum MAE value for both groups was for middle and front of tibia, then top of the shoe. This study demonstrates that top of the shoe is the best sensor location for GED and GRF prediction.

Detecting Gait Events from Accelerations Using Reservoir Computing.

Segmenting the gait cycle into multiple phases using gait event detection (GED) is a well-researched subject with many accurate algorithms. However, the algorithms that are able to perform accurate and robust GED for real-life environments and physical diseases tend to be too complex for their implementation on simple hardware systems limited in computing power and memory, such as those used in wearable devices. This study focuses on a numerical implementation of a reservoir computing (RC) algorithm called the echo state network (ESN) that is based on simple computational steps that are easy to implement on portable hardware systems for real-time detection. RC is a neural network method that is widely used for signal processing applications and uses a fast-training method based on a ridge regression adapted to the large quantity and variety of IMU data needed to use RC in various real-life environment GED. In this study, an ESN was used to perform offline GED with gait data from IMU and ground force sensors retrieved from three databases for a total of 28 healthy adults and 15 walking conditions. Our main finding is that despite its low complexity, ESN is robust for GED, with performance comparable to other state-of-the-art algorithms. Our results show the ESN is robust enough to obtain good detection results in all conditions if the algorithm is trained with variable data that match those conditions. The distribution of the mean absolute errors (MAE) between the detection times from the ESN and the force sensors were between 40 and 120 ms for 6 defined gait events (95th percentile). We compared our ESN with four different state-of-the-art algorithms from the literature. The ESN obtained a MAE not more than 10 ms above three other reference algorithms for normal walking indoor and outdoor conditions and yielded the 2nd lowest MAE and the 2nd highest true positive rate and specificity when applied to outdoor walking and running conditions. Our work opens the door to using the ESN as a GED for applications in wearable sensors for long-term patient monitoring.

Computing with networks of nonlinear mechanical oscillators.

As it is getting increasingly difficult to achieve gains in the density and power efficiency of microelectronic computing devices because of lithographic techniques reaching fundamental physical limits, new approaches are required to maximize the benefits of distributed sensors, micro-robots or smart materials. Biologically-inspired devices, such as artificial neural networks, can process information with a high level of parallelism to efficiently solve difficult problems, even when implemented using conventional microelectronic technologies. We describe a mechanical device, which operates in a manner similar to artificial neural networks, to solve efficiently two difficult benchmark problems (computing the parity of a bit stream, and classifying spoken words). The device consists in a network of masses coupled by linear springs and attached to a substrate by non-linear springs, thus forming a network of anharmonic oscillators. As the masses can directly couple to forces applied on the device, this approach combines sensing and computing functions in a single power-efficient device with compact dimensions.

Massive Mutagenesis: high-throughput combinatorial site-directed mutagenesis.

Massive Mutagenesis is a proprietary library creation method that enables the fast generation of high-quality genetic libraries. Starting from a single gene on a plasmid and hundreds to thousands of oligonucleotides, a one-step single-strand circular amplification method creates random combinations of several site-directed substitutions, insertions, or deletions. Libraries of up to a billion such variants have been routinely generated. Sequencing those variants demonstrated lower biases than alternative approaches such as error-prone PCR. Screening and selecting them has yielded improved biocatalysts, therapeutic proteins, and antibodies.

High-throughput site-directed mutagenesis using oligonucleotides synthesized on DNA chips.

Site-directed mutagenesis has greatly helped researchers both to understand the precise role of specific residues in coding sequences and to generate variants of proteins that have acquired new characteristics. Today's demands for more complete functional cartographies of proteins and advances in selection and screening technologies require that site-directed mutagenesis be adapted for high-throughput applications. We describe here the first generation of a library of single and multiple site-directed mutants using a mixture of oligonucleotides synthesized on DNA chips. We have used the human interleukin 15 (IL15) gene as a model, of which 37 codons were simultaneously targeted for substitution by any of eight possible codons. Ninety-six clones were sequenced, exhibiting a broad spectrum of targeted substitutions over the whole gene length with no unwanted mutations. Libraries produced using such pools of oligonucleotides open new perspectives to direct the evolution of proteins in vitro, by enabling the simple, rapid, and cost-effective generation of large tailor-made genetic diversities from any gene.

Long mRNAs coding for yeast mitochondrial proteins of prokaryotic origin preferentially localize to the vicinity of mitochondria.

BACKGROUND: Subcellular messenger RNA localization is important in most eukaryotic cells, even in unicellular organisms like yeast for which this process has been underestimated. Microarrays are rarely used to study subcellular mRNA localization at whole-genome level, but can be adapted to that purpose. This work focuses on studying the repartition of yeast nuclear transcripts encoding mitochondrial proteins between free cytosolic polysomes and polysomes bound to the mitochondrial outer membrane. RESULTS: Combining biochemical fractionations with oligonucleotide array analyses permits clustering of genes on the basis of the subcellular sites of their mRNA translation. A large fraction of yeast nuclear transcripts known to encode mitochondrial proteins is found in mitochondrial outer-membrane-bound fractions. These results confirm and extend a previous analysis conducted with partial genomic microarrays. Interesting statistical relations among mRNA localization, gene origin and mRNA lengths were found: longer and older mRNAs are more prone to be localized to the vicinity of mitochondria. These observations are included in a refined model of mitochondrial protein import. CONCLUSIONS: Mitochondrial biogenesis requires concerted expression of the many genes whose products make up the organelle. In the absence of any clear transcriptional program, coordinated mRNA localization could be an important element of the time-course of organelle construction. We have built a 'MitoChip' localization database from our results which allows us to identify interesting genes whose mRNA localization might be essential for mitochondrial biogenesis in most eukaryotic cells. Moreover, many components of the experimental and data-analysis strategy implemented here are of general relevance in global transcription studies.

In Saccharomyces cerevisiae, ATP2 mRNA sorting to the vicinity of mitochondria is essential for respiratory function.

We recently demonstrated that polysome-associated mRNAs that co-isolate with mitochondria encode a subset of mitochondrial proteins, and that the 3' UTRs of these transcripts are essential for their localization to the vicinity of the organelle. To address the question of the involvement of the mRNA targeting process in mitochondrial biogenesis, we studied the role of ATP2 3' UTR. An altered ATP2 allele in which the 3' UTR was replaced by the ADH1 3' UTR exhibits properties supporting the importance of mRNA localization to the vicinity of mitochondria: (i) the mutated strain presents a respiratory dysfunction; (ii) mitochondrial import of the protein translated from the altered gene is strongly reduced, even though the precursor is addressed to the organelle surface; (iii) systematic deletions of ATP2 3' UTR revealed a 100 nucleotide element presenting RNA targeting properties. Additionally, when the ATM1 3' UTR was replaced by the ADH1 3' UTR, we obtained cells in which ATM1 mRNA is also delocalized, and presenting a respiratory dysfunction. This demonstrates that mRNA localization to the vicinity of mitochondria plays a critical role in organelle biogenesis.