selecting-multiple-biomarker-subsets-with-similarly-effective-binary

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances

Biomarker detection is one of the more important biomedical questions for high-throughput &#39;omics&#39; researchers, and almost all existing biomarker ...

Jilin University

Neural stem cells (NSCs) are usually affected by a number of biological functions in neural traumatic and degenerative diseases. Autophagy may be involved in these diseases. However, whether autophagy could affect NSCs is largely unknown. Therefore, we aimed to investigate intracellular microstructures, proliferation, axon extension, and Beclin-1 expression of rat NSCs by basic culture medium and conditioned medium without epidermal growth factor (EGF) and basic fibroblast growth factor (bFGF). Transmission electron microscopy showed the formation of autophagic vacuoles in conditioned medium. Compared to the control group with normal medium, the number of secondary neurosphere was significantly reduced whereas the expression of Beclin-1 was enhanced. The majority of NSCs were nestin-positive when EGF/bFGF was withdrawn for 3 days and showed enhanced neurite extension, which was suppressed by autophagy antagonist 3-methyladenine. Our findings revealed that a short-term paucity of mitogens in microenvironments could induce autophagy of NSCs, which facilitated NSCs' axonal growth.

Autophagy induced by the withdrawal of mitogens promotes neurite extension in rat neural stem cells.

In a previous study, we reported a persistent reduction of F-actin puncta but a compensating increase in puncta size in the mouse hippocampus at 2 months after pilocarpine-induced status epilepticus (Epilepsy Res. 108 (2014), 379-389). However, the F-actin changes during the period of epileptogenesis remain unknown. This study was designed to examine the temporal and spatial changes of F-actin during the period of epileptogenesis in a pilocarpine-induced epilepsy model, primarily by the histological and TUNEL evaluation of cell loss, phalloidin detection of F-actin, and immunohistochemical analysis of glial reactions. The results demonstrated that F-actin continued to decrease after pilocarpine treatment, which was consistent in its time course with hippocampal neuronal death. Within different hippocampal subfields, the spatial changes of F-actin exhibited similar features. First, the F-actin puncta decreased in number. Thereafter, F-actin was transiently aggregated in dendritic shafts and neuronal cell bodies and eventually was completely lost in the degenerated neurons. The progressive changes of F-actin in the degenerating neurons reported in this study may help to elucidate a cytoskeletal mechanism that may link to the delayed cell loss that occurs during epileptogenesis. 

The progressive changes of filamentous actin cytoskeleton in the hippocampal neurons after pilocarpine-induced status epilepticus.

Cervical carcinoma is one of the leading causes of cancer-related death for female worldwide. Eukaryotic initiation factor 5A2 belongs to the eukaryotic initiation factor 5A family and is proposed to be a key factor involved in the development of diverse cancers. In the current study, a series of in vivo and in vitro investigations were performed to characterize the role of eukaryotic initiation factor 5A2 in oncogenesis and metastasis of cervical carcinoma. The expression status of eukaryotic initiation factor 5A2 in 15 cervical carcinoma patients was quantified. Then, the effect of eukaryotic initiation factor 5A2 knockdown on in vivo tumorigenicity ability, cell proliferation, cell cycle distribution, and cell mobility of HeLa cells was measured. To uncover the mechanism driving the function of eukaryotic initiation factor 5A2 in cervical carcinoma, expression of members within RhoA/ROCK pathway was detected, and the results were further verified with an RhoA overexpression modification. The level of eukaryotic initiation factor 5A2 in cervical carcinoma samples was significantly higher than that in paired paratumor tissues ( P < .05). And the in vivo tumorigenic ability of HeLa cells was reduced by inhibition of eukaryotic initiation factor 5A2. Knockdown of eukaryotic initiation factor 5A2 in HeLa cells decreased the cell viability compared with normal cells and induced G1 phase cell cycle arrest ( P < .05). Moreover, the cell migration ability of eukaryotic initiation factor 5A2 knockdown cells was dramatically inhibited. Associated with alterations in phenotypes, RhoA, ROCK I, and ROCK II were downregulated. The above-mentioned changes in eukaryotic initiation factor 5A2 knockdown cells were alleviated by the overexpression of RhoA. The major findings outlined in the current study confirmed the potential of eukaryotic initiation factor 5A2 as a promising prognosis predictor and therapeutic target for cervical carcinoma treatment. Also, our data inferred that eukaryotic initiation factor 5A2 might function in carcinogenesis of cervical carcinoma through an RhoA/ROCK-dependent manner.

Blocking Modification of Eukaryotic Initiation 5A2 Antagonizes Cervical Carcinoma via Inhibition of RhoA/ROCK Signal Transduction Pathway.

Binary classification is a widely employed problem to facilitate the decisions on various biomedical big data questions, such as clinical drug trials between treated participants and controls, and genome-wide association studies (GWASs) between participants with or without a phenotype. A machine learning model is trained for this purpose by optimizing the power of discriminating samples from two groups. However, most of the classification algorithms tend to generate one locally optimal solution according to the input dataset and the mathematical presumptions of the dataset. Here we demonstrated from the aspects of both disease classification and feature selection that multiple different solutions may have similar classification performances. So the existing machine learning algorithms may have ignored a horde of fishes by catching only a good one. Since most of the existing machine learning algorithms generate a solution by optimizing a mathematical goal, it may be essential for understanding the biological mechanisms for the investigated classification question, by considering both the generated solution and the ignored ones.

Multiple similarly-well solutions exist for biomedical feature selection and classification problems.

A correction to this article has been published and is linked from the HTML version of this paper. The error has been fixed in the paper.

Author Correction: Multiple similarly effective solutions exist for biomedical feature selection and classification problems.

Transcriptomic and methylomic patterns represent two major OMIC data sources impacted by both inheritable genetic information and environmental factors, and have been widely used as disease diagnosis and prognosis biomarkers.

An OMIC biomarker detection algorithm TriVote and its application in methylomic biomarker detection.

A correction to this article has been published and is linked from the HTML and PDF versions of this paper. The error has been fixed in the paper.

Publisher Correction: Multiple similarly effective solutions exist for biomedical feature selection and classification problems.

To investigate the effects of progranulin on migration and invasion of human colon cancer cells.

Jiamei Liu

College of Software

Jilin University

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances