Representation Learning and Pattern Recognition in Cognitive Biometrics: A Survey.

Cognitive biometrics is an emerging branch of biometric technology. Recent research has demonstrated great potential for using cognitive biometrics in versatile applications, including biometric recognition and cognitive and emotional state recognition. There is a major need to summarize the latest developments in this field. Existing surveys have mainly focused on a small subset of cognitive biometric modalities, such as EEG and ECG. This article provides a comprehensive review of cognitive biometrics, covering all the major biosignal modalities and applications. A taxonomy is designed to structure the corresponding knowledge and guide the survey from signal acquisition and pre-processing to representation learning and pattern recognition. We provide a unified view of the methodological advances in these four aspects across various biosignals and applications, facilitating interdisciplinary research and knowledge transfer across fields. Furthermore, this article discusses open research directions in cognitive biometrics and proposes future prospects for developing reliable and secure cognitive biometric systems.

Deep Learning in Diverse Intelligent Sensor Based Systems.

Deep learning has become a predominant method for solving data analysis problems in virtually all fields of science and engineering. The increasing complexity and the large volume of data collected by diverse sensor systems have spurred the development of deep learning methods and have fundamentally transformed the way the data are acquired, processed, analyzed, and interpreted. With the rapid development of deep learning technology and its ever-increasing range of successful applications across diverse sensor systems, there is an urgent need to provide a comprehensive investigation of deep learning in this domain from a holistic view. This survey paper aims to contribute to this by systematically investigating deep learning models/methods and their applications across diverse sensor systems. It also provides a comprehensive summary of deep learning implementation tips and links to tutorials, open-source codes, and pretrained models, which can serve as an excellent self-contained reference for deep learning practitioners and those seeking to innovate deep learning in this space. In addition, this paper provides insights into research topics in diverse sensor systems where deep learning has not yet been well-developed, and highlights challenges and future opportunities. This survey serves as a catalyst to accelerate the application and transformation of deep learning in diverse sensor systems.

Functional lipidomics: Palmitic acid impairs hepatocellular carcinoma development by modulating membrane fluidity and glucose metabolism.

Lipids are essential cellular components and energy sources of living organisms, and altered lipid composition is increasingly recognized as a signature of cancer. We performed lipidomic analysis in a series of hepatocellular carcinoma (HCC) cells and identified over 1,700 intact lipids originating from three major lipid categories. Comparative lipidomic screening revealed that 93 significantly changed lipids and decreased palmitic acyl (C16:0)-containing glycerophospholipids were positively associated with metastatic abilities of HCC cells. Furthermore, both in vitro and in vivo experiments demonstrated that C16:0 incubation specifically reduced malignant cell proliferation, impaired cell invasiveness, and suppressed tumor growth in mouse xenograft models. Biochemical experiments demonstrated that C16:0 treatment decreased cell membrane fluidity and limited glucose metabolism. A phosphoproteomics approach further revealed such C16:0 incubation attenuated phosphorylation levels of mammalian target of rapamycin (mTOR) and signal transducer and activator of transcription 3 (STAT3) pathway proteins. Multiple reaction monitoring analysis of 443 lipid molecules showed 8 reduced C16:0-containing lipids out of total 10 altered lipids when cancer tissues were compared with adjacent nontumor tissues in a cohort of clinical HCC specimens (P < 0.05). CONCLUSION: These data collectively demonstrate the biomedical potential of using altered lipid metabolism as a diagnostic marker for cancerous cells and open an opportunity for treating aggressive HCCs by targeting altered C16:0 metabolism. (Hepatology 2017;66:432-448).

Systematic comparison between SDS-PAGE/RPLC and high-/low-pH RPLC coupled tandem mass spectrometry strategies in a whole proteome analysis.

SDS-PAGE and high-pH RPLC are commonly used fractionation strategies in proteomics research. A comparative investigation of these two strategies would be meaningful to thoroughly understand their respective features. Here, we systematically compared the two methods by trying 4 SDS-PAGE/RPLC and 3 high-/low-pH RPLC different workflows for a higher sensitivity in protein identification. Totally 9793 proteins were identified in HepG2 cells, with 8581 proteins identified by high-/low-pH RPLC workflows and 7933 by SDS-PAGE/RPLC workflows. The results demonstrate that using high-pH RPLC in the first dimensional separation would favour a high-throughput proteome analysis but choosing SDS-PAGE could yield much better peptide coverage. We found that the SDS-PAGE fractionation method benefits the neutral pI peptides. We also analyzed unexpected modifications caused by the two strategies. Our results suggest that more pre-fractionation benefits protein identifications in both strategies and pooling of gel pieces according to their grey values increased the identification efficiency in SDS-PAGE/RPLC workflows.

Rapid and sensitive profiling and quantification of the human cell line proteome by LC-MS/MS without prefractionation.

In this paper, we demonstrate a rapid and reproducible 1D LC-MS/MS workflow for fast quantitative proteomic research. We have optimized the LC-MS/MS conditions, including digestion and gradient conditions, sample loading amount, and MS parameter settings. As a result, we were able to obtain twice as many protein identifications compared with the LC-MS/MS conditions before optimization. More than 4500 protein groups and 50 000 peptides were identified in less than 8 h without any fractionation. This 1D workflow was then applied to the analysis of the MLN4924 treated/untreated human umbilical vein endothelial cell (HUVEC) samples with label-free quantification. In these experiments, a total of 179 proteins showed a statistically significant expression change after the MLN4924 treatment. Functional analysis showed that these proteins are associated with cell death and survival; gene expression; cell cycle; and DNA replication, recombination, and repair.

Omics evidence: single nucleotide variants transmissions on chromosome 20 in liver cancer cell lines.

Cancer genomics unveils many cancer-related mutations, including some chromosome 20 (Chr.20) genes. The mutated messages have been found in the corresponding mRNAs; however, whether they could be translated to proteins still requires more evidence. Herein, we proposed a transomics strategy to profile the expression status of human Chr.20 genes (555 in Ensembl v72). The data of transcriptome and translatome (the mRNAs bound with ribosome, translating mRNAs) revealed that ∼80% of the coding genes on Chr.20 were detected with mRNA signals in three liver cancer cell lines, whereas of the proteome identified, only ∼45% of the Chr.20 coding genes were detected. The high amount of overlapping of identified genes in mRNA and RNC-mRNA (ribosome nascent-chain complex-bound mRNAs, translating mRNAs) and the consistent distribution of the abundance averages of mRNA and RNC-mRNA along the Chr.20 subregions in three liver cancer cell lines indicate that the mRNA information is efficiently transmitted from transcriptional to translational stage, qualitatively and quantitatively. Of the 457 genes identified in mRNAs and RNC-mRNA, 136 were found to contain SNVs with 213 sites, and >40% of these SNVs existed only in metastatic cell lines, suggesting them as the metastasis-related SNVs. Proteomics analysis showed that 16 genes with 20 SNV sites were detected with reliable MS/MS signals, and some SNVs were further validated by the MRM approach. With the integration of the omics data at the three expression phases, therefore, we are able to achieve the overall view of the gene expression of Chr.20, which is constructive in understanding the potential trend of encoding genes in a cell line and exploration of a new type of markers related to cancers.

Chromosome-8-coded proteome of Chinese Chromosome Proteome Data set (CCPD) 2.0 with partial immunohistochemical verifications.

We upgraded the preliminary CCPD 1.0 to CCPD 2.0 using the latest deep-profiling proteome (CCPD 2013) of three hepatocellular carcinoma (HCC) cell lines, namely, Hep3B, MHCC97H, and HCCLM3 (ProteomeXchange identifiers: PXD000529, PXD000533, and PXD000535). CCPD 2.0 totally covered 63.6% (438/689) of Chr. 8-coded proteins and 62.6% (439/701) of Chr. 8-coded protein-coding genes. Interestingly, we found that the missing proteins exhibited a tendency to form a cluster region in chromosomes, such as two ß-defensins clusters in Chr. 8, caused perhaps by their inflammation-related features. For the 41 Chr. 8-coded proteins being weakly or barely identified previously, we have performed an immunohistochemical (IHC) verification in 30 pairs of carcinoma/para-carcinoma HCC and 20 noncancerous liver tissues and confirmed their expressional evidence and occurrence proportions in tissue samples. We also verified 13 Chr. 8-coded HCC tumorigenesis-associated depleting or deficient proteins reported in CCPD 1.0 using IHC and screened 16 positive and 24 negative HCC metastatic potential-correlated proteins from large-scale label-free proteome quantitation data of CCPD 2013. Our results suggest that the selection of proper samples and the methodology to look for targeted missing proteins should be carefully considered in further verifications for the remaining Chr. 8-coded proteins.

A Compound Loss Function With Shape Aware Weight Map for Microscopy Cell Segmentation.

Microscopy cell segmentation is a crucial step in biological image analysis and a challenging task. In recent years, deep learning has been widely used to tackle this task, with promising results. A critical aspect of training complex neural networks for this purpose is the selection of the loss function, as it affects the learning process. In the field of cell segmentation, most of the recent research in improving the loss function focuses on addressing the problem of inter-class imbalance. Despite promising achievements, more work is needed, as the challenge of cell segmentation is not only the inter-class imbalance but also the intra-class imbalance (the cost imbalance between the false positives and false negatives of the inference model), the segmentation of cell minutiae, and the missing annotations. To deal with these challenges, in this paper, we propose a new compound loss function employing a shape aware weight map. The proposed loss function is inspired by Youden's J index to handle the problem of inter-class imbalance and uses a focal cross-entropy term to penalize the intra-class imbalance and weight easy/hard samples. The proposed shape aware weight map can handle the problem of missing annotations and facilitate valid segmentation of cell minutiae. Results of evaluations on all ten 2D+time datasets from the public cell tracking challenge demonstrate 1) the superiority of the proposed loss function with the shape aware weight map, and 2) that the performance of recent deep learning-based cell segmentation methods can be improved by using the proposed compound loss function.

FingerGAN: A Constrained Fingerprint Generation Scheme for Latent Fingerprint Enhancement.

Latent fingerprint enhancement is an essential preprocessing step for latent fingerprint identification. Most latent fingerprint enhancement methods try to restore corrupted gray ridges/valleys. In this paper, we propose a new method that formulates latent fingerprint enhancement as a constrained fingerprint generation problem within a generative adversarial network (GAN) framework. We name the proposed network FingerGAN. It can enforce its generated fingerprint (i.e, enhanced latent fingerprint) indistinguishable from the corresponding ground truth instance in terms of the fingerprint skeleton map weighted by minutia locations and the orientation field regularized by the FOMFE model. Because minutia is the primary feature for fingerprint recognition and minutia can be retrieved directly from the fingerprint skeleton map, we offer a holistic framework that can perform latent fingerprint enhancement in the context of directly optimizing minutia information. This will help improve latent fingerprint identification performance significantly. Experimental results on two public latent fingerprint databases demonstrate that our method outperforms the state of the arts significantly. The codes will be available for non-commercial purposes from https://github.com/HubYZ/LatentEnhancement.

[Inactivation of Cd and As by an Enterobacter Isolated from Cd and As Contaminated Farmland Soil].

A strain of Enterobacter was screened from cadmium and arsenic contaminated farmland soil and its passivation mechanism of cadmium and arsenic were explored through removing performance and characterization experiments. The results showed that the screened strain M5 was identified as Enterobacter sp. with a sulfate-reduction function, and its maximum resistance concentration was approximately 1 mmol·L-1 to cadmium and arsenic. In the simulation system, the maximum removal efficiencies of cadmium and arsenic were 94.13% and 27.26% by strain M5, respectively. The results of SEM-EDS and XRD confirmed that Cd and As were fixed to CdS and As2S3, and XPS results showed that carboxyl groups, hydroxyl groups, and amide groups on the surface of the bacteria were mainly involved in biological adsorption. These results can provide new ideas and a theoretical basis for microbial applications to soil remediations for heavy metal pollution.

[Effect of Combined Application of an Enterobacter and Sulfur Fertilizer on Cadmium and Arsenic Accumulation in Rice].

The aim of this study was to explore the effects of different sulfur fertilizers combined with sulfate-reducing bacteria on the accumulation of cadmium and arsenic in rice and the formation of iron plaque under long-term flooding conditions and to provide a reference for the safe production of rice fields polluted by moderate and mild cadmium and arsenic. We adopted a pot experiment, selecting two sulfur fertilizers, sulfur and calcium sulfate, and Enterobacter M5 with sulfate-reducing ability, and designed six treatments of single application and combined application of different sulfur fertilizers and M5. The results showed that the combined application of calcium sulfate and M5(CM5) had the best effect on reducing available cadmium and arsenic in rice rhizosphere soil. The combined application of sulfur fertilizer or M5 could reduce the content of cadmium and inorganic arsenic in early season rice grains by 8%-51% and 42%-61%, respectively, under flooding conditions. The content of cadmium and inorganic arsenic in late rice grains decreased by 81%-92% and 41%-62%, respectively. The treatment of the combined application of sulfur and M5(SM5) and CM5 had the best effect on reducing cadmium and arsenic content in both early and late season rice grains. SM5 and CM5 could promote the adsorption of cadmium and arsenic by iron plaque, and the extracted cadmium and arsenic content of ACA in both treatments was significantly higher than that of CK. The extracted iron content of ACA in the CM5 treatment was also significantly higher than that of CK, which indicates that the combined application of calcium sulfate and M5 would promote the formation of iron plaque. The results showed that the combined application of sulfur fertilizer and M5 was better than single application in reducing the content of cadmium and arsenic in grains, whereas the combined application of calcium sulfate and M5 was the best and most stable method.

Quantitative proteomic analysis of serum proteins in patients with Parkinson's disease using an isobaric tag for relative and absolute quantification labeling, two-dimensional liquid chromatography, and tandem mass spectrometry.

Parkinson's disease (PD) is a common disease which occurs in aged people with chronic, progressive degenerative character of the central nervous system. Until now there is no effective treatment method in PD patients before they show obvious symptoms for prevention and early diagnosis. In order to find out early disease specific biomarkers, two-dimensional liquid chromatography-tandem mass spectrometry coupled with isobaric tags for relative and absolute quantification (iTRAQ) labeling was employed to quantitatively identify the differentially expressed proteins among the different disease progress types of PD. 26 proteins were differentially expressed in a total of 258 identified proteins by proteomic techniques. The expression level of eight proteins which included sero-transferrin and clusterin increased. The expression level of eighteen proteins which include complement component 4B, apolipoprotein A-I, α-2-antiplasmin and coagulation factor V decreased. Those proteins may be associated with oxidative stress, mitochondrial dysfunction, abnormal protein aggregation and inflammation. In this study, the expression level of apolipoprotein A-I decreased, particularly in the early stage of PD patients. This protein regulated not only the lipid metabolism in the central nervous system, but also influenced the deposition process of proteins which are involved in neural degenerative diseases, such as the pathogenesis of PD.

3D Fingerprint Recognition based on Ridge-Valley-Guided 3D Reconstruction and 3D Topology Polymer Feature Extraction.

An automated fingerprint recognition system (AFRS) for 3D fingerprints is essential and highly promising for biometric security. Despite the progress in developing 3D AFRSs, achieving high-quality real-time reconstruction and high-accuracy recognition of 3D fingerprints remain two challenging issues. To address them, we propose a robust 3D AFRS based on ridge-valley (RV)-guided 3D fingerprint reconstruction and 3D topology polymer (TTP) feature extraction. The former considers the unique fingerprint characteristics of the RV and achieves real-time reconstruction. Unlike traditional triangulation-based methods that establish correspondences between points by cross-correlation-based searching, we propose to establish RV correspondences (RVCs) between ridges/valleys by defining and calculating a RVC matrix based on the topology of RV curves. To enhance depth reconstruction, curve-based smoothing is proposed to refine our novel RV disparity map. The TTP feature codes the 3D topology by projecting the 3D minutiae onto multiple planes and extracting their corresponding 2D topologies and has proven to be effective and efficient for 3D fingerprint recognition. Comprehensive experimental results demonstrate that our method outperforms the state-of-the-art methods in terms of both reconstruction and recognition accuracy. Also, due to its very short running time, it is appropriate for practical applications.

In-tip nanoreactors for cancer cells proteome profiling.

Mass spectrometry (MS)-based proteome profiling is essential for molecular diagnostics in modern biomedical study. To date, sample preparation including protein extraction and proteolysis is still very challenging and lack of efficiency. Recently tips-based sample preparation protocols exhibit strong potentials to achieve the goal of "a proteome in an hour". However, in-tip proteolysis is still rarely reported and far from ideal for dealing with complex bio-samples. In this work, nanoreactors encapsulated micropipette tips were demonstrated as high performance devices for fast (∼minutes) and multiplexing proteolysis to assist the profiling of cancer cells proteome. Nanoporous silica materials with controlled pore size and surface chemistry were prepared as nanoreactors and encapsulated in micropipette tips for efficient in situ proteolysis. The as-constructed device showed desirable sensitivity (LOD of 0.204 ± 0.008 ng/µL and LOQ of 0.937 ± 0.055 ng/µL), selectivity, stability (two months under -20 °C), reusability (at least 10 times), and little memory effect in MS based bottom-up proteomic analysis. It was used for comprehensive protein mapping from cancer cell lines. The number of identified proteins was increased by 18%, 22%, 52%, and 52% dealing with HepG2, F56, MCF7, and HCCLM3 cancer cells, compared to traditional in-solution proteolysis based bottom-up proteomic strategy. With the enhanced performance, our work built a novel, efficient and miniaturized platform for facile proteomic sample preparation, which is promising for advanced biomarkers discovery in biomedical study.

Analgesic effects and possible mechanisms of iridoid glycosides from Lamiophlomis rotata (Benth.) Kudo in rats with spared nerve injury.

ETHNOPHARMACOLOGICAL RELEVANCE: Lamiophlomis rotata (Benth.) Kudo (L. rotata) is a medical plant that has been traditionally used for centuries for the treatment of pain, such as bone and muscle pain, joint pain and dysmenorrhea. Although iridoid glycosides of L. rotata (IGLR) are the major active components of it according to reports, it still remains poorly understood about the molecular mechanisms underlying analgesic effects of IGLR. The aim of the present study was to investigate the analgesic effect of IGLR on a spared nerve injury (SNI) model of neuropathic pain. MATERIALS AND METHODS: The SNI model in rats was established by complete transection of the common peroneal and tibial distal branches of the sciatic nerve, leaving the sural branch intact. Then SNI rats were treated with IGLR for 14 days, using normal saline as the negative control. The paw withdrawal mechanical threshold (PMWT) in response to mechanical stimulation was measured by von Frey filaments on day 1 before operation and on days 1, 3, 5, 7, 9, 11, 13 and 14 after operation, respectively. After 14 days, the levels of nitric oxide (NO), nitric oxide synthase (NOS), tumor necrosis factor-α (TNF-α), interleukin-1ß (IL-1ß), interleukin-10 (IL-10) and cyclic guanosine monophosphate (cGMP) in the spinal dorsal horn were measured by the corresponding kits, mRNA expression of inducible NOS (iNOS) and protein kinase G type I (PKGI) of spinal cord were analyzed by reverse-transcription polymerase chain reaction (RT-PCR). The expression of N-methyl-D-aspartate receptor (NMDAR) and protein kinase C (PKCγ) of the spinal dorsal horn was performed by Western blot. Before all the experiments, motor coordination performance and locomotor activity had been tested. RESULTS: Our results showed that remarkable mechanical allodynia was observed on day 1 after operation in the SNI model, which was accompanied by a decrease in PMWT. Treatment with IGLR (200, 400, 800mg/kg) significantly alleviated SNI-induced mechanical allodynia, markedly decreased the levels of NO, NOS, TNF-α, IL-1ß and cGMP, and increased the level of IL-10. Meanwhile, IGLR (200, 400, 800mg/kg) also inhibited the protein expression of NMDAR, PKCγ and the mRNA expression of iNOS and PKGΙ in the spinal cord. In addition, gavage with the IGLR aqueous extract (800mg/kg) did not signifiantly alter motor coordination or locomotor activity. CONCLUSIONS: These results indicated IGLR could produce an anti-neuropathic pain effect that might partly be related to the inhibition of the NO/cGMP/PKG and NMDAR/PKC pathways and the level of TNF-α, IL-1ß as well as to the increase of the level of IL-10 in spinal cord.

Large scale systematic proteomic quantification from non-metastatic to metastatic colorectal cancer.

A systematic proteomic quantification of formalin-fixed, paraffin-embedded (FFPE) colorectal cancer tissues from stage I to stage IIIC was performed in large scale. 1017 proteins were identified with 338 proteins in quantitative changes by label free method, while 341 proteins were quantified with significant expression changes among 6294 proteins by iTRAQ method. We found that proteins related to migration expression increased and those for binding and adherent decreased during the colorectal cancer development according to the gene ontology (GO) annotation and ingenuity pathway analysis (IPA). The integrin alpha 5 (ITA5) in integrin family was focused, which was consistent with the metastasis related pathway. The expression level of ITA5 decreased in metastasis tissues and the result has been further verified by Western blotting. Another two cell migration related proteins vitronectin (VTN) and actin-related protein (ARP3) were also proved to be up-regulated by both mass spectrometry (MS) based quantification results and Western blotting. Up to now, our result shows one of the largest dataset in colorectal cancer proteomics research. Our strategy reveals a disease driven omics-pattern for the metastasis colorectal cancer.

[A method using spectrum alignment to improve analysis efficiency of liquid chromatography combined with mass spectrometry].

In the analysis of proteins in human umbilical vein endothelial cells (HUVEC) treated with dimethyl sulfoxide (DMSO) and NEDD8-activating enzyme inhibitor (MLN4924, MLN), the Progenesis LC-MS software (Nonlinear Dynamics Ltd) was applied to liquid chromatography spectrum alignment, while spectrum similarities were figured out among several experiments of the same sample, and also among different samples. After double enzymolysis, the sample was added with digested QconCAT standard proteins. They were separated by HPLC-MS/MS, followed by spectrum alignment and data analysis. This established experiment flow offered a better identification result of more than 8 000 proteins, while the original result was about 7 000 proteins, ensuring a relatively high identification efficiency. On the basis of relative quantification with spectrum count, the described procedure can analyze the differential expression of proteins induced by DMSO and MLN. The similarities of total ion chromatograms after alignment were also compared. This method was proved to be quick and easy, with the advantages of high throughput and high sensitivity.