Signal Genetics (United States)

Consumer wearables and sensors are a rich source of data about patients’ daily disease and symptom burden, particularly in the case of movement disorders like Parkinson’s disease (PD). However, interpreting these complex data into so-called digital biomarkers requires complicated analytical approaches, and validating these biomarkers requires sufficient data and unbiased evaluation methods. Here we describe the use of crowdsourcing to specifically evaluate and benchmark features derived from accelerometer and gyroscope data in two different datasets to predict the presence of PD and severity of three PD symptoms: tremor, dyskinesia, and bradykinesia. Forty teams from around the world submitted features, and achieved drastically improved predictive performance for PD status (best AUROC = 0.87), as well as tremor- (best AUPR = 0.75), dyskinesia- (best AUPR = 0.48) and bradykinesia-severity (best AUPR = 0.95).

The growth and survival factor hepatocyte growth factor (HGF) is expressed at high levels in multiple myeloma (MM) cells. We report here that elevated HGF transcription in MM was traced to DNA mutations in the promoter alleles of HGF. Sequence analysis revealed a previously undiscovered single-nucleotide polymorphism (SNP) and crucial single-nucleotide variants (SNVs) in the promoters of myeloma cells that produce large amounts of HGF. The allele-specific mutations functionally reassembled wild-type sequences into the motifs that affiliate with endogenous transcription factors NFKB (nuclear factor kappa-B), MZF1 (myeloid zinc finger 1), and NRF-2 (nuclear factor erythroid 2-related factor 2). In vitro, a mutant allele that gained novel NFKB-binding sites directly responded to transcriptional signaling induced by tumor necrosis factor alpha (TNFα) to promote high levels of luciferase reporter. Given the recent discovery by genome-wide sequencing (GWS) of numerous non-coding mutations in myeloma genomes, our data provide evidence that heterogeneous SNVs in the gene regulatory regions may frequently transform wild-type alleles into novel transcription factor binding properties to aberrantly interact with dysregulated transcriptional signals in MM and other cancer cells.

Abstract Alkali metal silicides have ability to enhance oil recovery in a variety of light, medium and heavy oil reservoirs. These chemicals, which include the silicides of sodium (Na), potassium (K) and lithium (Li), are free-flowing granules or very fine powders that are applied downhole in hydrocarbon dispersions. When introduced into a formation through an appropriate non-aqueous carrier fluid, these materials rapidly react with the water in the reservoir pore space, releasing hydrogen gas and heat, and converting into alkali silicates. The silicide-water reaction combined with the flooding process provides multiple mechanisms in the reservoir to enhance oil recovery. Enhanced oil recovery mechanisms include: energy addition through the generation of hydrogen; oil viscosity reductions due to hydrogen solubilization, temperature increase and solvent dilution from the carrier fluid; interfacial tension reduction due to in-situ surfactant generation from interaction of the crude oil organic acids in the reservoir oil with the alkalinity from the produced silicates; and potential improvement of water wettability in carbonate reservoirs. This one chemical combines the effects of thermal, drive energy and chemical mechanisms. In this work, a field-scale numerical simulation study was conducted to investigate the feasibility of cyclically injecting an alkali metal silicide into the wormhole structures of a post CHOPS (cold heavy oil production with sand) reservoir. The Computer Modeling Group's (CMG) STARS simulator was used to perform the simulations; the model consists of six vertical wells with wormhole structures developed using proprietary wormhole growth models that are based on actual field production histories from a representative CHOPS field in Canada (the Lloydminster, Alberta and Saskatchewan region). Multiple simulation cases were run to investigate the effects of injected cycle volume, cycle time, injection rate and silicide concentration. A sensitivity analysis was performed on parameters affecting the slurry model and dispersion rates of the silicide in the reservoir. The preliminary economics of the process were calculated and used to identify an optimized and cost-effective injection strategy, which can subsequently be used as a basis to design a field trial application. The study results showed that the cyclic injection of sodium silicide in a post CHOPS reservoir can in fact improve the recovery of oil in place. The study shows that the predominant recovery mechanism is likely the pressure maintenance of the reservoir that provides energy for continued oil production. This, coupled with secondary oil viscosity reductions, enable the cyclic injection of silicide to increase production for an additional 5 to 10 years, thereby adding 25 to 50% to recoverable reserves under favorable economics. This paper discusses the impacts of the in situ generation of heat, hydrogen and alkali silicate for post CHOPS augmentation and summarizes the key findings of the simulation study and economic modeling.

Bistatic image feature-based space-time processing (BIFSTP) is proposed for effective ground moving target detection for airborne bistatic radars. Moving target detection from ground clutter using BIFSTP is performed based on the separation of target and clutter and their different features in the Doppler-angle domain without the need for clutter estimation. The proposed algorithm is simulated for typical airborne bistatic radar operation geometries.

Traditionally the radiation beams of phased array antennas are formed by linearly shifting the phases of the transmitted waveforms from the antenna array elements, and the antenna beams can be scanned from one direction to another by subsequently changing the waveform phases [1, 2]. Digitally forming receiving antenna beams can be implemented by digitally filtering the received signals at array elements and have been widely used. However, digitally forming virtual transmitting beams or both transmitting and receiving beams has been very technically challenging. Yet there are many potential advantages for employing virtual transmitting beams rather than real phased array antenna beams. For example, in forming virtual transmitting antenna beams, the radiation energy of the antenna is not physically focused in any particular direction and thus is very difficult to be detected. As a result, the virtual radiating system can operate in a dasiasilentpsila mode with an extremely low probability of interception (LPI). Because the virtual radiation beam is formed through digital filtering of the received signals, the formed beams are more accurate and concentrated for intended radiation direction compared with phased array antenna beams. In addition, with virtual radiation systems, multiple transmitting-receiving beams can be generated simultaneously. In this work, we propose an innovative approach to forming virtual digital antenna beams (VDAB) by transmitting and processing orthogonal coding waveforms. The principle and architecture of the proposed system will be introduced and some simulated results will be presented to demonstrate the feasibility of the new systems.

We consider in this paper a likelihood principle based approach for the range dependent problem in space time adaptive processing. The proposed generalized likelihood ratio test (GLRT) addresses the range dependent issue by directly applying the likelihood principle to the range dependent signal model. Using the knowledge of platform geometry, we develop maximum likelihood estimators that facilitate the GLRT. This differs from existing methods that rely on data transformations in dealing with the range dependence issue. Numerical examples show that the new GLRT approach exhibits significant performance gain over existing approaches.