HIV-1 usurps mixed-charge domain-dependent CPSF6 phase separation for higher-order capsid binding, nuclear entry and viral DNA integration.

HIV-1 integration favors nuclear speckle (NS)-proximal chromatin and viral infection induces the formation of capsid-dependent CPSF6 condensates that colocalize with nuclear speckles (NSs). Although CPSF6 displays liquid-liquid phase separation (LLPS) activity in vitro, the contributions of its different intrinsically disordered regions, which includes a central prion-like domain (PrLD) with capsid binding FG motif and C-terminal mixed-charge domain (MCD), to LLPS activity and to HIV-1 infection remain unclear. Herein, we determined that the PrLD and MCD both contribute to CPSF6 LLPS activity in vitro. Akin to FG mutant CPSF6, infection of cells expressing MCD-deleted CPSF6 uncharacteristically arrested at the nuclear rim. While heterologous MCDs effectively substituted for CPSF6 MCD function during HIV-1 infection, Arg-Ser domains from related SR proteins were largely ineffective. While MCD-deleted and wildtype CPSF6 proteins displayed similar capsid binding affinities, the MCD imparted LLPS-dependent higher-order binding and co-aggregation with capsids in vitro and in cellulo. NS depletion reduced CPSF6 puncta formation without significantly affecting integration into NS-proximal chromatin, and appending the MCD onto a heterologous capsid binding protein partially restored virus nuclear penetration and integration targeting in CPSF6 knockout cells. We conclude that MCD-dependent CPSF6 condensation with capsids underlies post-nuclear incursion for viral DNA integration and HIV-1 pathogenesis.

Human CCR6+ Th Cells Show Both an Extended Stable Gradient of Th17 Activity and Imprinted Plasticity.

Th17 cells have been investigated in mice primarily for their contributions to autoimmune diseases. However, the pathways of differentiation of Th17 and related Th cells (type 17 cells) and the structure of the type 17 memory population in humans are not well understood; such understanding is critical for manipulating these cells in vivo. By exploiting differences in levels of surface CCR6, we found that human type 17 memory cells, including individual T cell clonotypes, form an elongated continuum of type 17 character along which cells can be driven by increasing RORγt. This continuum includes cells preserved within the memory pool with potentials that reflect the early preferential activation of multiple over single lineages. The phenotypes and epigenomes of CCR6+ cells are stable across cell divisions under noninflammatory conditions. Nonetheless, activation in polarizing and nonpolarizing conditions can yield additional functionalities, revealing, respectively, both environmentally induced and imprinted mechanisms that contribute differentially across the type 17 continuum to yield the unusual plasticity ascribed to type 17 cells.

CD47 and IFT57 Are Colinear Genes That Are Highly Coexpressed in Most Cancers and Exhibit Parallel Cancer-Specific Correlations with Survival.

An association between high CD47 expression and poor cancer survival has been attributed to its function on malignant cells to inhibit phagocytic clearance. However, CD47 mRNA expression in some cancers lacks correlation or correlates with improved survival. IFT57 encodes an essential primary cilium component and is colinear with CD47 across amniote genomes, suggesting coregulation of these genes. Analysis of The Cancer Genome Atlas datasets identified IFT57 as a top coexpressed gene with CD47 among 1156 human cancer cell lines and in most tumor types. The primary cilium also regulates cancer pathogenesis, and correlations between IFT57 mRNA and survival paralleled those for CD47 in thyroid and lung carcinomas, melanoma, and glioma. CD47 ranked first for coexpression with IFT57 mRNA in papillary thyroid carcinomas, and higher expression of both genes correlated with significantly improved overall survival. CD47 and IFT57 mRNAs were coordinately regulated in thyroid carcinoma cell lines. Transcriptome analysis following knockdown of CD47 or IFT57 in thyroid carcinoma cells identified the cytoskeletal regulator CRACD as a specific target of IFT57. CRACD mRNA expression inversely correlated with IFT57 mRNA and with survival in low-grade gliomas, lung adenocarcinomas, and papillary thyroid carcinomas, suggesting that IFT57 rather than CD47 regulates survival in these cancers.

CD47-Dependent Regulation of Immune Checkpoint Gene Expression and MYCN mRNA Splicing in Murine CD8 and Jurkat T Cells.

Elevated expression of CD47 in some cancers is associated with poor survival related to its function as an innate immune checkpoint when expressed on tumor cells. In contrast, elevated CD47 expression in cutaneous melanomas is associated with improved survival. Previous studies implicated protective functions of CD47 expressed by immune cells in the melanoma tumor microenvironment. RNA sequencing analysis of responses induced by CD3 and CD28 engagement on wild type and CD47-deficient Jurkat T lymphoblast cells identified additional regulators of T cell function that were also CD47-dependent in mouse CD8 T cells. MYCN mRNA expression was upregulated in CD47-deficient cells but downregulated in CD47-deficient cells following activation. CD47 also regulated alternative splicing that produces two N-MYC isoforms. The CD47 ligand thrombospondin-1 inhibited expression of these MYCN mRNA isoforms, as well as induction of the oncogenic decoy MYCN opposite strand (MYCNOS) RNA during T cell activation. Analysis of mRNA expression data for melanomas in The Cancer Genome Atlas identified a significant coexpression of MYCN with CD47 and known regulators of CD8 T cell function. Thrombospondin-1 inhibited the induction of TIGIT, CD40LG, and MCL1 mRNAs following T cell activation in vitro. Increased mRNA expression of these T cell transcripts and MYCN in melanomas was associated with improved overall survival.

Amylases from thermophilic bacteria: structure and function relationship.

Amylases hydrolyze starch to diverse products including dextrins and progressively smaller polymers of glucose units. Thermally stable amylases account for nearly 25% of the enzyme market. This review highlights the structural attributes of the α-amylases from thermophilic bacteria. Heterologous expression of amylases in suitable hosts is discussed in detail. Further, specific value maximization approaches, such as protein engineering and immobilization of the amylases are discussed in order to improve its suitability for varied applications on a commercial scale. The review also takes into account of the immobilization of the amylases on nanomaterials to increase the stability and reusability of the enzymes. The function-based metagenomics would provide opportunities for searching amylases with novel characteristics. The review is expected to explore novel amylases for future potential applications.

CXCR3 Identifies Human Naive CD8+ T Cells with Enhanced Effector Differentiation Potential.

In mice, the ability of naive T (TN) cells to mount an effector response correlates with TCR sensitivity for self-derived Ags, which can be quantified indirectly by measuring surface expression levels of CD5. Equivalent findings have not been reported previously in humans. We identified two discrete subsets of human CD8+ TN cells, defined by the absence or presence of the chemokine receptor CXCR3. The more abundant CXCR3+ TN cell subset displayed an effector-like transcriptional profile and expressed TCRs with physicochemical characteristics indicative of enhanced interactions with peptide-HLA class I Ags. Moreover, CXCR3+ TN cells frequently produced IL-2 and TNF in response to nonspecific activation directly ex vivo and differentiated readily into Ag-specific effector cells in vitro. Comparative analyses further revealed that human CXCR3+ TN cells were transcriptionally equivalent to murine CXCR3+ TN cells, which expressed high levels of CD5. These findings provide support for the notion that effector differentiation is shaped by heterogeneity in the preimmune repertoire of human CD8+ T cells.

Genetic and Phenotypic Heterogeneity of the Nocardiopsis alba Strains of Seawater.

This study deals with the genetic and phenotypic heterogeneity of the marine Nocardiopsis alba strains isolated during pre-monsoon, monsoon and post-monsoon seasons. The isolates were characterized for their morphological and biochemical attributes, growth media preferences, antibiotic susceptibility and extracellular enzyme secretion. Nocardiopsis alba strains were assessed against 12 different antibiotics, and the responses were expressed in terms of the multiple antibiotic resistance (MAR) number. The majority of the strains produced multiple extracellular enzymes: proteases, amylases and lipases. Further, the strains were characterized on the basis of 16S rRNA gene sequencing and the majority were identified as Nocardiopsis alba along with few strains of Streptomyces lopnurensis, Nocardiopsis synnemataformans and Nocardiopsis dassonvillei. Neighbor-joining (NJ) phylogenetic tree suggested variation among the genetically similar Nocardiopsis alba species. The study establishes significant heterogeneity with respect to genetic and phenotypic characteristics of the strains of Nocardiopsis alba. Phylogenetic tree and phenogram-based comparison reflect the heterogeneity in terms of different clustering patterns of the strains. Further, the whole genome sequence data available in the literature also confirm the observed heterogeneity. Nocardiopsis alba strains displayed a relatively regressive pattern of dependence on the environmental factors based on the canonical correspondence analysis plot. The study represents cultivation, characterization, phylogenetic analysis and enzymatic potential of the Nocardiopsis alba species of seawater origin.

Infrared Monitoring of Oxygenation Process Generated by Robotic Verticalization in Bedridden People.

Bedridden people, especially at home, suffer from diverse pathologies beyond the main one that brings them to a specific position. Long-term cares are suitable at home to avoid congestions within hospital facilities. There are different technologies available to improve such people's conditions in their daily life. The standing posture is the key solution to enhance people's wellness amid the psychological burden due to the almost impossibility to be completely healed. The paper proposes the use of a polyfunctional and robotic bed capable of displaying many positions namely vertical, tilting, anti-trendelenburg with necessary graduation. A three-year monitoring of a patient, using a polyfunctional and robotic bed, suffering from amyotrophic lateral sclerosis (ALS), has been investigated. Different physiological parameters have been measured and, particularly, the variation of temperature has been measured in presence of body position connected to the robotic bed rotation that provokes biomechanical effort. It is demonstrated that certain body positions correspond to major and minor physical effort, hence major and minor oxygenation. An infrared camera has been used. As a positive result, the variation of posture has been delaying the increase of the pathological signs, because of better conditions.

3D Deep Learning on Medical Images: A Review.

The rapid advancements in machine learning, graphics processing technologies and the availability of medical imaging data have led to a rapid increase in the use of deep learning models in the medical domain. This was exacerbated by the rapid advancements in convolutional neural network (CNN) based architectures, which were adopted by the medical imaging community to assist clinicians in disease diagnosis. Since the grand success of AlexNet in 2012, CNNs have been increasingly used in medical image analysis to improve the efficiency of human clinicians. In recent years, three-dimensional (3D) CNNs have been employed for the analysis of medical images. In this paper, we trace the history of how the 3D CNN was developed from its machine learning roots, we provide a brief mathematical description of 3D CNN and provide the preprocessing steps required for medical images before feeding them to 3D CNNs. We review the significant research in the field of 3D medical imaging analysis using 3D CNNs (and its variants) in different medical areas such as classification, segmentation, detection and localization. We conclude by discussing the challenges associated with the use of 3D CNNs in the medical imaging domain (and the use of deep learning models in general) and possible future trends in the field.

Image Thresholding Improves 3-Dimensional Convolutional Neural Network Diagnosis of Different Acute Brain Hemorrhages on Computed Tomography Scans.

Intracranial hemorrhage is a medical emergency that requires urgent diagnosis and immediate treatment to improve patient outcome. Machine learning algorithms can be used to perform medical image classification and assist clinicians in diagnosing radiological scans. In this paper, we apply 3-dimensional convolutional neural networks (3D CNN) to classify computed tomography (CT) brain scans into normal scans (N) and abnormal scans containing subarachnoid hemorrhage (SAH), intraparenchymal hemorrhage (IPH), acute subdural hemorrhage (ASDH) and brain polytrauma hemorrhage (BPH). The dataset used consists of 399 volumetric CT brain images representing approximately 12,000 images from the National Neuroscience Institute, Singapore. We used a 3D CNN to perform both 2-class (normal versus a specific abnormal class) and 4-class classification (between normal, SAH, IPH, ASDH). We apply image thresholding at the image pre-processing step, that improves 3D CNN classification accuracy and performance by accentuating the pixel intensities that contribute most to feature discrimination. For 2-class classification, the F1 scores for various pairs of medical diagnoses ranged from 0.706 to 0.902 without thresholding. With thresholding implemented, the F1 scores improved and ranged from 0.919 to 0.952. Our results are comparable to, and in some cases, exceed the results published in other work applying 3D CNN to CT or magnetic resonance imaging (MRI) brain scan classification. This work represents a direct application of a 3D CNN to a real hospital scenario involving a medically emergent CT brain diagnosis.

Desertibacillus haloalkaliphilus gen. nov., sp. nov., isolated from a saline desert.

Two Gram-stain-positive, rod-shaped and endospore-forming bacteria that represent a single species, designated strains KJ1-10-99T and KJ1-10-93, were isolated from a saline desert of Little Rann of Kutch, Gujarat, India. Analysis of 16S rRNA gene sequences revealed that the isolates belonged to the family Bacillaceae and were closely related to each other with 16S rRNA gene sequence similarity of 99.9 %. However, these two isolates formed a novel phylogenetic branch within this family. Both strains were aerobic, catalase and oxidase positive, and could grow optimally at 37 °C and pH 9. Further, strains KJ1-10-99T and KJ1-10-93 grew optimally at a NaCl concentration of 7.5 and 15 % (w/v), respectively. Both strains shared highest sequence similarity with Fermentibacillus polygoni IEB3T (96.90 %) followed by Bacillus nanhaiisediminis NH3T (96.3 %) and Bacillus alkalinitrilicus ANL-iso4T (96.3 %). The major cellular fatty acids were anteiso-C15 : 0, anteiso-C17:0, C16 : 0, and iso-C15 : 0. The major polar lipids were diphosphatidylglycerol and phosphatidylglycerol in both strains. The predominant isoprenoid quinone was MK-7 in both the strains. The peptidoglycan contained meso-diaminopimelic acid (meso-DAP) as the diagnostic diamino acid. The DNA G+C content of strains KJ1-10-99T and KJ1-10-93 were 48.7 and 48.9 mol% respectively. Both strains could be distinguished from closest phylogenetic neighbours based on a number of phenotypic properties. On the basis of polyphasic taxonomic analysis and phylogenetic data, we conclude that the strains KJ1-10-99T (=LMG 29918T=KCTC 33878T) and KJ1-10-93 (=LMG 29919=KCTC 33877) represent a novel species of a new genus in the family Bacillaceae, order Bacillales, for which the name Desertibacillus haloalkaliphilus gen. nov., sp. nov. is proposed.

PLZF regulates CCR6 and is critical for the acquisition and maintenance of the Th17 phenotype in human cells.

Th17 cells, which express the chemokine receptor CCR6, are implicated in many immune-mediated disorders, such as psoriasis and multiple sclerosis. We found that expression levels of CCR6 on human effector/memory CD4(+) T cells reflect a continuum of Th17 differentiation. By evaluating the transcriptome in cells with increasing CCR6, we detected progressive upregulation of ZBTB16, which encodes the broad complex, tramtrack, bric-à-brac-zinc finger transcription factor promyelocytic leukemia zinc finger protein (PLZF). Using chromatin immunoprecipitation for modified histones, p300, and PLZF, we identified enhancer-like sites at -9/-10 and -13/-14 kb from the upstream transcription start site of CCR6 that bind PLZF in CCR6(+) cells. For Th cells from adult blood, both in the CCR6(+) memory population and in naive cells activated ex vivo, knockdown of ZBTB16 downregulated CCR6 and other Th17-associated genes. ZBTB16 and RORC (which encodes the "master regulator" RORγt) cross-regulate each other, and PLZF binds at the RORC promoter in CCR6(+) cells. In naive Th cells from cord blood, ZBTB16 expression was confined to CD161(+) cells, which are Th17 cell precursors. ZBTB16 was not expressed in mouse Th17 cells, and Th17 cells could be made from luxoid mice, which harbor an inactivating mutation in Zbtb16. These studies demonstrate a role for PLZF as an activator of transcription important both for Th17 differentiation and the maintenance of the Th17 phenotype in human cells, expand the role of PLZF as a critical regulator in the human adaptive immune system, and identify a novel, essential element in a regulatory network that is of significant therapeutic interest.

Conventional NK cells can produce IL-22 and promote host defense in Klebsiella pneumoniae pneumonia.

It was reported that host defense against pulmonary Klebsiella pneumoniae infection requires IL-22, which was proposed to be of T cell origin. Supporting a role for IL-22, we found that Il22(-/-) mice had decreased survival compared with wild-type mice after intratracheal infection with K. pneumoniae. Surprisingly, however, Rag2(-/-) mice did not differ from wild-type mice in survival or levels of IL-22 in the lungs postinfection with K. pneumoniae. In contrast, K. pneumoniae-infected Rag2(-/-)Il2rg(-/-) mice failed to produce IL-22. These data suggested a possible role for NK cells or other innate lymphoid cells in host defense and production of IL-22. Unlike NK cell-like innate lymphoid cells that produce IL-22 and display a surface phenotype of NK1.1(-)NKp46(+)CCR6(+), lung NK cells showed the conventional phenotype, NK1.1(+)NKp46(+)CCR6(-). Mice depleted of NK cells using anti-asialo GM1 showed decreased survival and higher lung bacterial counts, as well as increased dissemination of K. pneumoniae to blood and liver, compared with control-treated mice. NK cell depletion also led to decreased production of IL-22 in the lung. Within 1 d postinfection, although there was no increase in the number of lung NK cells, a subset of lung NK cells became competent to produce IL-22, and such cells were found in both wild-type and Rag2(-/-) mice. Our data suggest that, during pulmonary infection of mice with K. pneumoniae, conventional NK cells are required for optimal host defense, which includes the production of IL-22.

CD47 signaling regulates the immunosuppressive activity of VEGF in T cells.

Thrombospondin-1 (TSP1) inhibits angiogenesis, in part, by interacting with the ubiquitous cell-surface receptor CD47. In endothelial cells, CD47 interacts directly with vascular endothelial growth factor receptor (VEGFR)-2, and TSP1 inhibits VEGFR2 phosphorylation and signaling by disrupting this association. We show that CD47 similarly associates with and regulates VEGFR2 in T cells. TSP1 inhibits phosphorylation of VEGFR2 and its downstream target Src in wild type but not in CD47-deficient human Jurkat and primary murine T cells. VEGFR2 signaling inhibits proliferation and TCR signaling in wild type T cells. However, ligation of CD47 by TSP1 or loss of CD47 expression reverses some inhibitory effects of VEGF on proliferation and T cell activation. We further found that VEGF and VEGFR2 expression are upregulated in CD47-deficient murine CD4(+) and human Jurkat T cells, and the resulting autocrine VEGFR2 signaling enhances proliferation and some TCR responses in the absence of CD47. Thus, CD47 signaling modulates the ability of VEGF to regulate proliferation and TCR signaling, and autocrine production of VEGF by T cells contributes to this regulation. This provides a mechanism to understand the context-dependent effects of TSP1 and VEGF on T cell activation, and reveals an important role for CD47 signaling in regulating T cell production of the major angiogenic factor VEGF.

An Improved CAD System for Breast Cancer Diagnosis Based on Generalized Pseudo-Zernike Moment and Ada-DEWNN Classifier.

In this paper, a novel framework of computer-aided diagnosis (CAD) system has been presented for the classification of benign/malignant breast tissues. The properties of the generalized pseudo-Zernike moments (GPZM) and pseudo-Zernike moments (PZM) are utilized as suitable texture descriptors of the suspicious region in the mammogram. An improved classifier- adaptive differential evolution wavelet neural network (Ada-DEWNN) is proposed to improve the classification accuracy of the CAD system. The efficiency of the proposed system is tested on mammograms from the Mammographic Image Analysis Society (mini-MIAS) database using the leave-one-out cross validation as well as on mammograms from the Digital Database for Screening Mammography (DDSM) database using 10-fold cross validation. The proposed method on MIAS-database attains a fair accuracy of 0.8938 and AUC of 0.935 (95 % CI = 0.8213-0.9831). The proposed method is also tested for in-plane rotation and found to be highly rotation invariant. In addition, the proposed classifier is tested and compared with some well-known existing methods using receiver operating characteristic (ROC) analysis using DDSM- database. It is concluded the proposed classifier has better area under the curve (AUC) (0.9289) and highly précised with 95 % CI, 0.8216 to 0.9834 and 0.0384 standard error.

Selectivity in the Use of Gi/o Proteins Is Determined by the DRF Motif in CXCR6 and Is Cell-Type Specific.

CXCR6, the receptor for CXCL16, is expressed on multiple cell types and can be a coreceptor for human immunodeficiency virus 1. Except for CXCR6, all human chemokine receptors contain the D(3.49)R(3.50)Y(3.51) sequence, and all but two contain A(3.53) at the cytoplasmic terminus of the third transmembrane helix (H3C), a region within class A G protein-coupled receptors that contacts G proteins. In CXCR6, H3C contains D(3.49)R(3.50)F(3.51)I(3.52)V(3.53) at positions 126-130. We investigated the importance and interdependence of the canonical D126 and the noncanonical F128 and V130 in CXCR6 by mutating D126 to Y, F128 to Y, and V130 to A singly and in combination. For comparison, we mutated the analogous positions D142, Y144, and A146 to Y, F, and V, respectively, in CCR6, a related receptor containing the canonical sequences. Mutants were analyzed in both human embryonic kidney 293T and Jurkat E6-1 cells. Our data show that for CXCR6 and/or CCR6, mutations in H3C can affect both receptor signaling and chemokine binding; noncanonical H3C sequences are functionally linked, with dual changes mitigating the effects of single mutations; mutations in H3C that compromise receptor activity show selective defects in the use of individual Gi/o proteins; and the effects of mutations in H3C on receptor function and selectivity in Gi/o protein use can be cell-type specific. Our findings indicate that the ability of CXCR6 to make promiscuous use of the available Gi/o proteins is exquisitely dependent on sequences within the H3C and suggest that the native sequence allows for preservation of this function across different cellular environments.

The blood-brain barrier is disrupted in a mouse model of infantile neuronal ceroid lipofuscinosis: amelioration by resveratrol.

Disruption of the blood-brain barrier (BBB) is a serious complication frequently encountered in neurodegenerative disorders. Infantile neuronal ceroid lipofuscinosis (INCL) is a devastating childhood neurodegenerative lysosomal storage disorder caused by palmitoyl-protein thioesterase-1 (PPT1) deficiency. It remains unclear whether BBB is disrupted in INCL and if so, what might be the molecular mechanism(s) of this complication. We previously reported that the Ppt1-knockout (Ppt1-KO) mice that mimic INCL manifest high levels of oxidative stress and neuroinflammation. Recently, it has been reported that CD4(+) T-helper 17 (T(H)17) lymphocytes may mediate BBB disruption and neuroinflammation, although the precise molecular mechanism(s) remain unclear. We sought to determine: (i) whether the BBB is disrupted in Ppt1-KO mice, (ii) if so, do T(H)17-lymphocytes underlie this complication, and (iii) how might T(H)17 lymphocytes breach the BBB. Here, we report that the BBB is disrupted in Ppt1-KO mice and that T(H)17 lymphocytes producing IL-17A mediate disruption of the BBB by stimulating production of matrix metalloproteinases (MMPs), which degrade the tight junction proteins essential for maintaining BBB integrity. Importantly, dietary supplementation of resveratrol (RSV), a naturally occurring antioxidant/anti-inflammatory polyphenol, markedly reduced the levels of T(H)17 cells, IL-17A and MMPs, and elevated the levels of tight junction proteins, which improved the BBB integrity in Ppt1-KO mice. Intriguingly, we found that RSV suppressed the differentiation of CD4(+) T lymphocytes to IL-17A-positive T(H)17 cells. Our findings uncover a mechanism by which T(H)17 lymphocytes mediate BBB disruption and suggest that small molecules such as RSV that suppress T(H)17 differentiation are therapeutic targets for neurodegenerative disorders such as INCL.

Extraction of the metagenomic DNA and assessment of the bacterial diversity from the petroleum-polluted sites.

The assessment of the microbial diversity of the entire community of a given habitat requires the extraction of the total environmental DNA. Metagenomic investigations of a petroleum-polluted habitat have its unique challenges. The specific methods were developed for the extraction of high-quality metagenome in good quantity from the petroleum-polluted saline and non-saline sites in Gujarat (India). The soil samples were washed to remove the toxic, hazardous organic pollutants which might interfere with the recovery of the metagenomic DNA. The metagenomic DNA extraction results were encouraging with the mechanical bead beating, soft lysis, and combination of both. The extracted DNA was assessed for its purity and yield followed by its application in the amplification of the 16S rRNA region. The amplicons were used for judging the molecular diversity by the denaturing gradient gel electrophoresis (DGGE). The microbial diversity was also analyzed statistically by calculating various diversity indices and principal component analysis (PCA). The results on the metagenomic diversity of the bacterial population among the three cohorts based on the culture-independent technique exhibited significant difference among the PAH sites and Okha-Madhi and Porbandar Madhavpur habitats.

Sparse Deep Neural Network for Encoding and Decoding the Structural Connectome.

Brain state classification by applying deep learning techniques on neuroimaging data has become a recent topic of research. However, unlike domains where the data is low dimensional or there are large number of available training samples, neuroimaging data is high dimensional and has few training samples. To tackle these issues, we present a sparse feedforward deep neural architecture for encoding and decoding the structural connectome of the human brain. We use a sparsely connected element-wise multiplication as the first hidden layer and a fixed transform layer as the output layer. The number of trainable parameters and the training time is significantly reduced compared to feedforward networks. We demonstrate superior performance of this architecture in encoding the structural connectome implicated in Alzheimer's disease (AD) and Parkinson's disease (PD) from DTI brain scans. For decoding, we propose recursive feature elimination (RFE) algorithm based on DeepLIFT, layer-wise relevance propagation (LRP), and Integrated Gradients (IG) algorithms to remove irrelevant features and thereby identify key biomarkers associated with AD and PD. We show that the proposed architecture reduces 45.1% and 47.1% of the trainable parameters compared to a feedforward DNN with an increase in accuracy by 2.6 % and 3.1% for cognitively normal (CN) vs AD and CN vs PD classification, respectively. We also show that the proposed RFE method leads to a further increase in accuracy by 2.1% and 4% for CN vs AD and CN vs PD classification, while removing approximately 90% to 95% irrelevant features. Furthermore, we argue that the biomarkers (i.e., key brain regions and connections) identified are consistent with previous literature. We show that relevancy score-based methods can yield high discriminative power and are suitable for brain decoding. We also show that the proposed approach led to a reduction in the number of trainable network parameters, an increase in classification accuracy, and a detection of brain connections and regions that were consistent with earlier studies.