Channel Input Adaptation via Natural Type Selection

The rolling bearing is the crucial component in the rotating machinery. The degradation process monitoring and remaining useful life prediction of the bearing are necessary for the condition-based maintenance. The commonly used deep learning methods use the raw or processed time domain data as the input. However, the feature extracted by these approaches is insufficient and incomprehensive. To tackle this problem, this paper proposed an improved Deep Convolution Neural Network with the dual-channel input from the time and frequency domain in parallel. The proposed methodology consists of two stages: the incipient failure identification and the degradation process fitting. To verify the effectiveness of the method, the IEEE PHM 2012 dataset is adopted to compare the proposed method and other commonly used approaches. The results show that the improved Deep Convolution Neural Network can effectively describe the degradation process for the rolling bearing.

Download Full-text

Variance analysis of immunoglobulin alleles in natural populations of rabbit (Oryctolagus cuniculus): the extensive interallelic divergence at the b locus could be the outcome of overdominance-type selection.

Genetics ◽

10.1093/genetics/135.1.171 ◽

1993 ◽

Vol 135 (1) ◽

pp. 171-187 ◽

Cited By ~ 1

Author(s):

W van der Loo

Keyword(s):

Heavy Chain ◽

Light Chain ◽

Gene Diversity ◽

Natural Populations ◽

Constant Region ◽

Immunoglobulin Light Chain ◽

Evolutionary Patterns ◽

B Locus ◽

Type Selection ◽

Predominant Allele

Abstract Population genetic data are presented which should contribute to evaluation of the hypothesis that the extraordinary evolutionary patterns observed at the b locus of the rabbit immunoglobulin light chain constant region can be the outcome of overdominance-type selection. The analysis of allele correlations in natural populations revealed an excess of heterozygotes of about 10% at the b locus while heterozygote excess was not observed at loci determining the immunoglobulin heavy chain. Data from the published literature, where homozygote advantage was suggested, were reevaluated and found in agreement with data here presented. Gene diversity was evenly distributed among populations and showed similarities with patterns reported for histocompatibility loci. Analysis of genotypic disequilibria revealed strong digenic associations between the leading alleles of heavy and light chain constant region loci in conjunction with trigenic disequilibria corresponding to a preferential association of b locus heterozygosity with the predominant allele of the heavy chain e locus. It is argued that this may indicate compensatory or nonadditive aspects of a putative heterozygosity enhancing mechanism, implying that effects at the light chain might be more pronounced in populations fixed for the heavy chain polymorphism.

Download Full-text

BIOM-60. BIOMARKER EVALUATION FOR IMIPRIDONE ONC206 REVEALS ClpP, ATF4, MYC, EGFR and HIF1 AS KEY PREDICTORS OF ANTI-CANCER EFFICACY

Neuro-Oncology ◽

10.1093/neuonc/noaa215.057 ◽

2020 ◽

Vol 22 (Supplement_2) ◽

pp. ii14-ii15

Author(s):

Sara Morrow ◽

Mathew Garnett ◽

Ultan McDermott ◽

Cyril Benes ◽

Joshua Allen ◽

...

Keyword(s):

Clinical Trials ◽

Strong Predictor ◽

Predictive Biomarkers ◽

Receptor Expression ◽

Tumor Type ◽

Anticancer Efficacy ◽

Type Selection ◽

Egfr Expression ◽

Transport Chain ◽

Biomarker Evaluation

Abstract ONC206 is a DRD2 antagonist and ClpP agonist that is a chemical derivative of ONC201, which is in Phase II clinical trials for H3 K27M-mutant glioma. We have previously reported that dopamine receptor expression correlates with ONC201 and ONC206 efficacy. Here, we evaluated additional predictive biomarkers for both agents in the GDSC panel using RNA-seq expression data. ClpP emerged as a strong predictor of efficacy for ONC206 (IC50: p = 1.83E-4, AUC: 9.92E-13). ClpP activation enhances degradation of its substrates, including electron transport chain members responsible for oxidative phosphorylation (OXPHOS). Imipridones activate the integrated stress response involving ATF4 and cause proteasomal degradation of c-myc, which also reduces OXPHOS. Accordingly, high ATF4 (IC50: p = 4.92E-5, AUC: p=2.84E-9) and elevated c-myc correlated with ONC206 efficacy (IC50: p = 6.42E-6, AUC: 3.31E-12). EGFR expression is inversely correlate with DRD2 expression in glioma and its activation is associated with glycolysis. Low EGFR expression correlated with increased ONC206 efficacy (IC50: p = 4.32E-7, AUC: p = 6.44E-20). Loss of HIF1 shunts cellular metabolism toward OXPHOS. Low HIF1 expression correlated with increased anticancer efficacy for ONC206 (IC50: p = 1.96E-3, AUC: p = 1.56E-11). Expression of each of the five markers was significantly different in cell lines that achieved an IC90 with ONC206 (~10%) versus those that did not. A similar analysis for ONC201 revealed that ClpP, MYC and EGFR are more predictive of efficacy relative to ATF4 and HIF1. Combinatorial biomarker analyses revealed MYC/EGFR as the most significant predictor of IC50 for both agents. ClpP/MYC and ClpP/HIF1 were the most significant predictors of AUC for ONC201 and ONC206, respectively. Ongoing studies are further investigating tumor type enrichment of biomarkers. Prediction of innate imipridone sensitivity using biomarkers identified in this study may guide patient and tumor type selection in clinical trials.

Download Full-text