RESUMO
Inherited retinal dystrophies (IRDs) are characterized by photoreceptor dysfunction or degeneration. Clinical and phenotypic overlap between IRDs makes the genetic diagnosis very challenging and comprehensive genomic approaches for accurate diagnosis are frequently required. While there are previous studies on IRDs in Pakistan, causative genes and variants are still unknown for a significant portion of patients. Therefore, there is a need to expand the knowledge of the genetic spectrum of IRDs in Pakistan. Here, we recruited 52 affected and 53 normal individuals from 15 consanguineous Pakistani families presenting non-syndromic and syndromic forms of IRDs. We employed single molecule Molecular Inversion Probes (smMIPs) based panel sequencing and whole genome sequencing to identify the probable disease-causing variants in these families. Using this approach, we obtained a 93% genetic solve rate and identified 16 (likely) causative variants in 14 families, of which seven novel variants were identified in ATOH7, COL18A1, MERTK, NDP, PROM1, PRPF8 and USH2A while nine recurrent variants were identified in CNGA3, CNGB1, HGSNAT, NMNAT1, SIX6 and TULP1. The novel MERTK variant and one recurrent TULP1 variant explained the intra-familial locus heterogeneity in one of the screened families while two recurrent CNGA3 variants explained compound heterozygosity in another family. The identification of variants in known disease-associated genes emphasizes the utilization of time and cost-effective screening approaches for rapid diagnosis. The timely genetic diagnosis will not only identify any associated systemic issues in case of syndromic IRDs, but will also aid in the acceleration of personalized medicine for patients affected with IRDs.
Assuntos
Consanguinidade , Sequenciamento de Nucleotídeos em Larga Escala , Linhagem , Humanos , Paquistão , Masculino , Feminino , Sequenciamento de Nucleotídeos em Larga Escala/métodos , Criança , Mutação , Adulto , Adolescente , Análise Mutacional de DNA , Adulto Jovem , Oftalmopatias Hereditárias/genética , Oftalmopatias Hereditárias/diagnóstico , Pré-Escolar , Distrofias Retinianas/genética , Distrofias Retinianas/diagnóstico , Testes Genéticos/métodos , Sequenciamento Completo do GenomaRESUMO
IDDBCS is a heterogeneous genetic syndrome with diverse clinical features including Intellectual disability and epilepsy. Using WES, Sanger sequencing, we identified a novel nonsense variant in the PHF21A gene responsible for IDDBCS syndrome. The patient has diverse and overlapping clinical phenotypes. The identified variant leads to abnormal secondary and tertiary structure of the protein and, consequently, affects its function.
Assuntos
Anormalidades Craniofaciais , Epilepsia , Deficiência Intelectual , Humanos , Deficiência Intelectual/genética , Epilepsia/genética , Síndrome , Fenótipo , Anormalidades Craniofaciais/genética , Histona Desacetilases/genéticaRESUMO
POPDC1 also known as BVES, is a highly conserved transmembrane protein, important for striated muscle function and homeostasis. Pathogenic variants in the POPDC1 gene are associated with limb-girdle muscular dystrophy type 25 (LGMDR25). In the present study, we performed trio-whole exome sequencing (WES) followed by Sanger sequencing on a single family having LGMD clinical features. Protein modeling of all POPDC1 missense variants (POPDC1Pro134Leu , POPDC1Ile193Ser , and POPDC1Ser201Phe ) associated with LGMDR25 were performed using Molecular Dynamics (MD) simulation. We identified a homozygous missense variant (c.401C>T; p.Pro134Leu) in the POPDC1 gene. Altered 3D structure, disruptive fluctuation, less compactness, and instability were observed in all the three variants of POPDC1 protein models. In comparison, POPDC1Ser201Phe protein dynamics were more unstable than other variants. Functional study of newly identified variant would add key answers to underlying mechanisms of the disease.
Assuntos
Distrofia Muscular do Cíngulo dos Membros , Humanos , Moléculas de Adesão Celular/genética , Homozigoto , Proteínas de Membrana/genética , Proteínas Musculares/genética , Músculo Esquelético/metabolismo , Distrofia Muscular do Cíngulo dos Membros/genética , Mutação de Sentido Incorreto/genéticaRESUMO
Infection of hepatitis C (HCV) is a major threat to human health throughout the world. The current therapy program suffers from restricted efficiency and low tolerance, and there is serious demand frr novel medication. NS3/4A protease is observed to be very effective target for the treatment of HCV. A data set of the already reported HCV NS3/4A protease inhibitors was first docked into the NS3/4A protease (PDB ID: 4A92A) active sites of both protease and helicase sites for calculating the docking score, binding affinity, binding mode, and solvation energy. Then the data set of these reported inhibitors was used in a computer-based program "RECAP Analyses" implemented in MOE to fragment every molecule in the subset according to simple retrosynthetic analysis rules. The RECAP analysis fragments were then used in another computer-based program "RECAP Synthesis" to randomly recombine and generate synthetically reasonable novel chemical structures. The novel chemical structures thus produced were then docked against HCV NS3/4A. After a thorough validation of all undertaken steps, based on Lipinski's rule of five, docking score, binding affinity, solvation energy, and Van der Waal's interactions with HCV NS3/4A, 12 novel chemical structures were identified as inhibitors of HCV NS3/4A. The novel structures thus designed are hoped to play a key role in the development of new effective inhibitors of HCV.
Assuntos
Hepatite C , Simulação de Dinâmica Molecular , Humanos , Endopeptidases/metabolismo , Hepacivirus , Hepatite C/tratamento farmacológico , Domínio Catalítico , Proteínas não Estruturais Virais/metabolismo , Inibidores de Proteases/química , Antivirais/químicaRESUMO
Background and Objective: Mutations in the CYB5R3 gene cause reduced NADH-dependent cytochrome b5 reductase enzyme function and consequently lead to recessive congenital methemoglobinemia (RCM). RCM exists as RCM type I (RCM1) and RCM type II (RCM2). RCM1 leads to higher methemoglobin levels causing only cyanosis, while in RCM2, neurological complications are also present along with cyanosis. Materials and Methods: In the current study, a consanguineous Pakistani family with three individuals showing clinical manifestations of cyanosis, chest pain radiating to the left arm, dyspnea, orthopnea, and hemoptysis was studied. Following clinical assessment, a search for the causative gene was performed using whole exome sequencing (WES) and Sanger sequencing. Various variant effect prediction tools and ACMG criteria were applied to interpret the pathogenicity of the prioritized variants. Molecular dynamic simulation studies of wild and mutant systems were performed to determine the stability of the mutant CYB5R3 protein. Results: Data analysis of WES revealed a novel homozygous missense variant NM_001171660.2: c.670A > T: NP_001165131.1: p.(Ile224Phe) in exon 8 of the CYB5R3 gene located on chromosome 22q13.2. Sanger sequencing validated the segregation of the identified variant with the disease phenotype within the family. Bioinformatics prediction tools and ACMG guidelines predicted the identified variant p.(Ile224Phe) as disease-causing and likely pathogenic, respectively. Molecular dynamics study revealed that the variant p.(Ile224Phe) in the CYB5R3 resides in the NADH domain of the protein, the aberrant function of which is detrimental. Conclusions: The present study expanded the variant spectrum of the CYB5R3 gene. This will facilitate genetic counselling of the same and other similar families carrying mutations in the CYB5R3 gene.
Assuntos
Metemoglobinemia , Humanos , Metemoglobinemia/congênito , Metemoglobinemia/genética , Simulação de Dinâmica Molecular , NAD/genética , NAD/metabolismo , Mutação , Cianose , Citocromo-B(5) Redutase/genética , Citocromo-B(5) Redutase/metabolismoRESUMO
BACKGROUND: Limb-girdle muscular dystrophy (LGMD) comprises a heterogeneous group of diseases, affecting different muscles, predominantly skeletal muscles and cardiac muscles of the body. LGMD is classified into two main subtypes A and B, which are further subclassified into eight dominant and thirty recessive subtypes. Three genes, namely POPDC1, POPDC2 and POPDC3, encode popeye domain-containing protein (POPDC), and the variants of POPDC1 and POPDC3 genes have been associated with LGMD. METHODS: In the present study, we performed whole-exome sequencing (WES) analysis on a single-family to investigate the hallmark features of LGMD. The results of WES were further confirmed by Sanger sequencing and 3D protein modeling was also conducted. RESULTS: WES data analysis and Sanger sequencing revealed a homozygous missense variant (c.460A>G; p.Lys154Glu) at a highly conserved amino acid position in the POPDC3. Mutations in the POPDC3 gene have been previously associated with recessive limb-girdle muscular dystrophy type 26. 3D protein modeling further suggested that the identified variant might affect the POPDC3 structure and proper function. CONCLUSIONS: The present study confirms the role of POPDC3 in LGMD, and will facilitate genetic counseling of the family to mitigate the risks of the carrier or affects on future pregnancies.
Assuntos
Moléculas de Adesão Celular , Proteínas Musculares , Distrofia Muscular do Cíngulo dos Membros , Moléculas de Adesão Celular/genética , Homozigoto , Humanos , Proteínas Musculares/genética , Músculo Esquelético , Distrofia Muscular do Cíngulo dos Membros/genética , Mutação , Mutação de Sentido IncorretoRESUMO
Parkinson's disease (PD) is the second most common and progressive neurodegenerative disease globally, with major symptoms like bradykinesia, impaired posture, and tremor. Several genetic and environmental factors have been identified but elucidating the main factors have been challenging due to the disease's complex nature. Diagnosis, prognosis, and management of such diseases are challenging and require effective targeted attention in developing countries. Recently, PD is growing rapidly in many crowded Asian countries as an alarming threat with inadequate knowledge of its prevalence, genetic architecture, and geographic distribution. This study gave an in-depth overview of the prevalence, incidence and genomic/genetics studies published so far in the Asian population. To the best of our knowledge, PD has increased significantly in several Asian countries, including China, South Korea, Japan, Thailand, and Israel over the past few years, requiring a greater level of care and attention. Genetic screening of families with PD at national levels and establishing an official database of PD cases are essential to get a comprehensive and conclusive view of the exact prevalence and genetic diversity of PD in the Asian population to properly manage and treat the disease.
Assuntos
Doença de Parkinson/etiologia , Ásia/epidemiologia , Biomarcadores , Suscetibilidade a Doenças , Predisposição Genética para Doença , Humanos , Incidência , Mutação , Doença de Parkinson/diagnóstico , Doença de Parkinson/epidemiologia , Doença de Parkinson/terapia , Avaliação de Resultados da Assistência ao Paciente , Fenótipo , Vigilância da População , PrevalênciaRESUMO
We propose to improve the visual object tracking by introducing a soft mask based low-level feature fusion technique. The proposed technique is further strengthened by integrating channel and spatial attention mechanisms. The proposed approach is integrated within a Siamese framework to demonstrate its effectiveness for visual object tracking. The proposed soft mask is used to give more importance to the target regions as compared to the other regions to enable effective target feature representation and to increase discriminative power. The low-level feature fusion improves the tracker robustness against distractors. The channel attention is used to identify more discriminative channels for better target representation. The spatial attention complements the soft mask based approach to better localize the target objects in challenging tracking scenarios. We evaluated our proposed approach over five publicly available benchmark datasets and performed extensive comparisons with 39 state-of-the-art tracking algorithms. The proposed tracker demonstrates excellent performance compared to the existing state-of-the-art trackers.
RESUMO
CNN-based trackers, especially those based on Siamese networks, have recently attracted considerable attention because of their relatively good performance and low computational cost. For many Siamese trackers, learning a generic object model from a large-scale dataset is still a challenging task. In the current study, we introduce input noise as regularization in the training data to improve generalization of the learned model. We propose an Input-Regularized Channel Attentional Siamese (IRCA-Siam) tracker which exhibits improved generalization compared to the current state-of-the-art trackers. In particular, we exploit offline learning by introducing additive noise for input data augmentation to mitigate the overfitting problem. We propose feature fusion from noisy and clean input channels which improves the target localization. Channel attention integrated with our framework helps finding more useful target features resulting in further performance improvement. Our proposed IRCA-Siam enhances the discrimination of the tracker/background and improves fault tolerance and generalization. An extensive experimental evaluation on six benchmark datasets including OTB2013, OTB2015, TC128, UAV123, VOT2016 and VOT2017 demonstrate superior performance of the proposed IRCA-Siam tracker compared to the 30 existing state-of-the-art trackers.
RESUMO
Fall induced damages are serious incidences for aged as well as young persons. A real-time automatic and accurate fall detection system can play a vital role in timely medication care which will ultimately help to decrease the damages and complications. In this paper, we propose a fast and more accurate real-time system which can detect people falling in videos captured by surveillance cameras. Novel temporal and spatial variance-based features are proposed which comprise the discriminatory motion, geometric orientation and location of the person. These features are used along with ensemble learning strategy of boosting with J48 and Adaboost classifiers. Experiments have been conducted on publicly available standard datasets including Multiple Cameras Fall (with 2 classes and 3 classes) and UR Fall Detection achieving percentage accuracies of 99.2, 99.25 and 99.0, respectively. Comparisons with nine state-of-the-art methods demonstrate the effectiveness of the proposed approach on both datasets.
Assuntos
Acidentes por Quedas/prevenção & controle , Processamento de Imagem Assistida por Computador/métodos , Área Sob a Curva , Humanos , Curva ROC , Máquina de Vetores de Suporte , Gravação em Vídeo , CaminhadaRESUMO
The phytochemical screening, antimicrobial, antioxidant and cytotoxic properties of Camellia sinensis were evaluated in the present study. The phytochemical screening revealed the presence of an applicable amount of lycopene, ß-carotenes, flavonoids and tannins in C. sinensis. Among the phytochemicals, tannin was found to be significantly higher in tea plant. The antimicrobial activity of plant extracts against selected bacterial strains namely, Escherichia coli, Staphylococcus aurous, Pseudomonas aeruginosa, Klebsiella pneumoniae, Acinetobacter baumannii, Marginella morganii and Haemophilus influenzae was investigated. The results showed that the stem part of C. sinensis presented greater antimicrobial potential than the leaf and root. Antioxidant activity (assessed through % inhibition of linoleic acid per oxidation test) was the highest (89.22%) in n-hexane extract of root part as compared to other extracts. Finally, the cytotoxicity analysis (haemolytic activity against human erythrocytes) of plant extract showed the negligible (%) lysis of RBCs ranging from 1.73 to 4.01%. In conclusion, it can be suggested that C. sinensis is the potential source to obtain bioactive phenolic compounds with high antimicrobial and antioxidant properties, which could possibly be exploited for the treatment of various infectious diseases.
Assuntos
Anti-Infecciosos/farmacologia , Antioxidantes/farmacologia , Camellia sinensis/química , Extratos Vegetais/farmacologia , Anti-Infecciosos/isolamento & purificação , Anti-Infecciosos/toxicidade , Antioxidantes/isolamento & purificação , Antioxidantes/toxicidade , Células Cultivadas , Eritrócitos/efeitos dos fármacos , Bactérias Gram-Negativas/efeitos dos fármacos , Bactérias Gram-Positivas/efeitos dos fármacos , Hemólise/efeitos dos fármacos , Humanos , Ácido Linoleico/química , Oxirredução , Extratos Vegetais/isolamento & purificação , Extratos Vegetais/toxicidade , Folhas de Planta/química , Raízes de Plantas/química , Caules de Planta/química , Solventes/químicaRESUMO
Over a decade ago, Pan et al. [IEEE TPAMI 25, 1552 (2003)] performed face recognition using only the spectral reflectance of the face at six points and reported around 95% recognition rate. Since their database is private, no one has been able to replicate these results. Moreover, due to the unavailability of public datasets, there has been no detailed study in the literature on the viability of facial spectral reflectance for person identification. In this study, we introduce a new public database of facial spectral reflectance profiles measured with a high precision spectrometer. For each of the 40 subjects, spectral reflectance was measured at the same six points as Pan et al. [IEEE TPAMI 25, 1552 (2003)] in multiple sessions and with time lapse. Furthermore, we sample the facial spectral reflectance from two public hyperspectral face image datasets and analyzed the data using state of the art face classification techniques. The best performing classifier achieved the maximum rank-1 identification rate of 53.8%. We conclude that facial spectral reflectance alone is not a reliable biometric for unconstrained face recognition.
RESUMO
The classification of gigapixel Whole Slide Images (WSIs) is an important task in the emerging area of computational pathology. There has been a surge of interest in deep learning models for WSI classification with clinical applications such as cancer detection or prediction of cellular mutations. Most supervised methods require expensive and labor-intensive manual annotations by expert pathologists. Weakly supervised Multiple Instance Learning (MIL) methods have recently demonstrated excellent performance; however, they still require large-scale slide-level labeled training datasets that require a careful inspection of each slide by an expert pathologist. In this work, we propose a fully unsupervised WSI classification algorithm based on mutual transformer learning. The instances (i.e., patches) from gigapixel WSIs are transformed into a latent space and then inverse-transformed to the original space. Using the transformation loss, pseudo labels are generated and cleaned using a transformer label cleaner. The proposed transformer-based pseudo-label generator and cleaner modules mutually train each other iteratively in an unsupervised manner. A discriminative learning mechanism is introduced to improve normal versus cancerous instance labeling. In addition to the unsupervised learning, we demonstrate the effectiveness of the proposed framework for weakly supervised learning and cancer subtype classification as downstream analysis. Extensive experiments on four publicly available datasets show better performance of the proposed algorithm compared to the existing state-of-the-art methods.
Assuntos
Algoritmos , Interpretação de Imagem Assistida por Computador , Humanos , Interpretação de Imagem Assistida por Computador/métodos , Aprendizado de Máquina não Supervisionado , Aprendizado Profundo , Neoplasias/diagnóstico por imagem , Processamento de Imagem Assistida por Computador/métodosRESUMO
In this paper, a rigorous theoretical study is conducted to analyze the influence of varying solvent compositions on the retention characteristics of elution profiles within a fixed-bed liquid chromatographic column. In gradient chromatography, the propagation speed of elution profiles is manipulated through a progressive variation in the mobile-phase composition. Consequently, enhanced separation of the mixture components can be achieved together with a reduction in the requisite recycling times for subsequent injections. In other words, both the efficiency and the selectivity of the column can be enhanced. The lumped kinetic model coupled with the convection-diffusion equation for the volume fraction of the solvent is applied to simulate the process. The resulting nonlinear model equations are numerically solved by applying a semidiscrete second-order finite-volume method. The numerical solutions are utilized to quantify the effects of gradient starting and ending times, solvent composition, solvent strength parameters, and gradient slope on the concentration profiles. Additionally, temporal numerical moments are plotted versus the starting and ending times of the gradient, and standard performance criteria are presented for evaluating the process performance. The outcomes of this investigation will contribute to further enhancements in gradient elution chromatography.
RESUMO
Lysosomal enzymes degrade cellular macromolecules, while their inactivation causes human hereditary metabolic disorders. Mucopolysaccharidosis IVA (MPS IVA; Moquio A syndrome) is one of the lysosomal storage disorders caused by a defective Galactosamine-6-sulfatase (GalN6S) enzyme. In several populations, disease incidence is elevated due to missense mutations brought on by non-synonymous allelic variation in the GalN6S enzyme. Here, we studied the effect of non-synonymous single nucleotide polymorphism (nsSNPs) on the structural dynamics of the GalN6S enzyme and its binding with N-acetylgalactosamine (GalNAc) using all-atom molecular dynamics simulation and an essential dynamics approach. Consequently, in this study, we have identified three functionally disruptive mutations in domain-I and domain-II, that is, S80L, R90W, and S162F, which presumably contribute to post-translational modifications. The study delineated that both domains work cooperatively, and alteration in domain II (S80L, R90W) leads to conformational changes in the catalytic site in domain-I, while mutation S162F mainly provokes higher residual flexibility of domain II. These results show that these mutations impair the hydrophobic core, implying that Morquio A syndrome is caused by misfolding of the GalN6S enzyme. The results also show the instability of the GalN6S-GalNAc complex upon substitution. Overall, the structural dynamics resulting from point mutations give the molecular rationale for Moquio A syndrome and, more importantly, the Mucopolysaccharidoses (MPS) family of diseases, re-establishing MPS IVA as a protein-folding disease.Communicated by Ramaswamy H. Sarma.
Assuntos
Mucopolissacaridose IV , Humanos , Mucopolissacaridose IV/genética , Acetilgalactosamina , Galactosamina , Dobramento de Proteína , SulfatasesRESUMO
The NOD-Like Receptor Protein-3 (NLRP3) inflammasome is a key therapeutic target for the treatment of epilepsy and has been reported to regulate inflammation in several neurological diseases. In this study, a machine learning-based virtual screening strategy has investigated candidate active compounds that inhibit the NLRP3 inflammasome. As machine learning-based virtual screening has the potential to accurately predict protein-ligand binding and reduce false positives outcomes compared to traditional virtual screening. Briefly, classification models were created using Support Vector Machine (SVM), Random Forest (RF), and K-Nearest Neighbor (KNN) machine learning methods. To determine the most crucial features of a molecule's activity, feature selection was carried out. By utilizing 10-fold cross-validation, the created models were analyzed. Among the generated models, the RF model obtained the best results as compared to others. Therefore, the RF model was used as a screening tool against the large chemical databases. Molecular operating environment (MOE) and PyRx software's were applied for molecular docking. Also, using the Amber Tools program, molecular dynamics (MD) simulation of potent inhibitors was carried out. The results showed that the KNN, SVM, and RF accuracy was 0.911 %, 0.906 %, and 0.946 %, respectively. Moreover, the model has shown sensitivity of 0.82 %, 0.78 %, and 0.86 % and specificity of 0.95 %, 0.96 %, and 0.98 % respectively. By applying the model to the ZINC and South African databases, we identified 98 and 39 compounds, respectively, potentially possessing anti-NLRP3 activity. Also, a molecular docking analysis produced ten ZINC and seven South African compounds that has comparable binding affinities to the reference drug. Moreover, MD analysis of the two complexes revealed that the two compounds (ZINC000009601348 and SANC00225) form stable complexes with varying amounts of binding energy. The in-silico studies indicate that both compounds most likely display their inhibitory effect by inhibiting the NLRP3 protein.
RESUMO
BACKGROUND: Optical coherence tomography (OCT) and near-infrared spectroscopy (NIRS) allow assessment of the anatomy (OCT) and composition (NIRS) of coronary lesions. We sought to examine the association between pre-stenting lipid core plaque (LCP), as assessed by NIRS and post-stenting thrombus formation, as assessed by OCT. METHODS: We reviewed the angiograms of nine patients who underwent coronary stenting in association with NIRS and OCT imaging. A large LCP by NIRS was defined as at least three 2-mm yellow blocks on the NIRS block chemogram with >200° angular extent. Intracoronary thrombus was defined as a mass of medium reflectivity protruding into the vessel lumen, discontinuous from the surface of the vessel wall. RESULTS: Mean age was 67 ± 7 years, and all patients were men, presenting with stable angina (56%), unstable angina (11%), or acute myocardial infarction (33%). The mean vessel lipid core burden index (LCBI) was 120 ± 45, and the mean highest 6-mm LCBI was 386 ± 190. Three patients had a large LCP and two of them (66%) developed intrastent thrombus after stent implantation compared to none of six patients without large LCPs (0%, P = 0.02). The thrombus resolved after intracoronary glycoprotein IIb/IIIa administration and balloon postdilation. Postprocedural myocardial infarction occurred in 33% versus 17% of patients with and without large LCP, respectively (P = 0.57). CONCLUSION: Stenting of large LCPs may be associated with intrastent thrombus formation, suggesting that more intensive anticoagulant and/or antiplatelet therapy may be beneficial in such lesions. © 2012 Wiley Periodicals, Inc.
Assuntos
Vasos Coronários/química , Oclusão de Enxerto Vascular/diagnóstico , Lipídeos/análise , Placa Aterosclerótica/química , Espectroscopia de Luz Próxima ao Infravermelho/métodos , Stents , Tomografia de Coerência Óptica/métodos , Idoso , Estenose Coronária/cirurgia , Vasos Coronários/patologia , Seguimentos , Oclusão de Enxerto Vascular/metabolismo , Humanos , Masculino , Intervenção Coronária Percutânea , Placa Aterosclerótica/patologia , Reprodutibilidade dos TestesRESUMO
Spectral Embedding (SE) has often been used to map data points from non-linear manifolds to linear subspaces for the purpose of classification and clustering. Despite significant advantages, the subspace structure of data in the original space is not preserved in the embedding space. To address this issue subspace clustering has been proposed by replacing the SE graph affinity with a self-expression matrix. It works well if the data lies in a union of linear subspaces however, the performance may degrade in real-world applications where data often spans non-linear manifolds. To address this problem we propose a novel structure-aware deep spectral embedding by combining a spectral embedding loss and a structure preservation loss. To this end, a deep neural network architecture is proposed that simultaneously encodes both types of information and aims to generate structure-aware spectral embedding. The subspace structure of the input data is encoded by using attention-based self-expression learning. The proposed algorithm is evaluated on six publicly available real-world datasets. The results demonstrate the excellent clustering performance of the proposed algorithm compared to the existing state-of-the-art methods. The proposed algorithm has also exhibited better generalization to unseen data points and it is scalable to larger datasets without requiring significant computational resources.
RESUMO
Automatic tissue classification is a fundamental task in computational pathology for profiling tumor micro-environments. Deep learning has advanced tissue classification performance at the cost of significant computational power. Shallow networks have also been end-to-end trained using direct supervision however their performance degrades because of the lack of capturing robust tissue heterogeneity. Knowledge distillation has recently been employed to improve the performance of the shallow networks used as student networks by using additional supervision from deep neural networks used as teacher networks. In the current work, we propose a novel knowledge distillation algorithm to improve the performance of shallow networks for tissue phenotyping in histology images. For this purpose, we propose multi-layer feature distillation such that a single layer in the student network gets supervision from multiple teacher layers. In the proposed algorithm, the size of the feature map of two layers is matched by using a learnable multi-layer perceptron. The distance between the feature maps of the two layers is then minimized during the training of the student network. The overall objective function is computed by summation of the loss over multiple layers combination weighted with a learnable attention-based parameter. The proposed algorithm is named as Knowledge Distillation for Tissue Phenotyping (KDTP). Experiments are performed on five different publicly available histology image classification datasets using several teacher-student network combinations within the KDTP algorithm. Our results demonstrate a significant performance increase in the student networks by using the proposed KDTP algorithm compared to direct supervision-based training methods.
RESUMO
The KRAS gene belongs to the RAS family and codes for 188 amino acid residues of KRAS protein, with a molecular mass of 21.6 kD. Non-synonymous single-nucleotide polymorphisms (nsSNPs) have been identified within the coding region in which some are associated with different diseases. However, structural changes are not well defined yet. In this study, we first categorized SNPs in the KRAS coding area and then used computational methods to determine their impact on the protein structure and stability. In addition, the three-dimensional model of KRAS was taken from the Protein Data Bank for structural modeling. Furthermore, genomic data were extracted from a variety of sources, including the 1000 Genome Project, dbSNPs, and ENSEMBLE, and assessed through in silico methods. Based on various tools used in this study, 10 out of 48 missense SNPs with rsIDs were found deleterious. The substitution of alanine for proline at position 146 pushed several residues toward the center of the protein. Arginine instead of leucine has a minor effect on protein structure and stability. In addition, the substitution of proline for leucine at the 34th position disrupted the structure and led to a bigger size than the wild-type protein, hence interrupting the protein interaction. Using the well-intended computational approach and applying several bioinformatic tools, we characterized and identified most damaging nsSNPs and further explored the structural dynamics and stability of KRAS protein.