Proteomic Characterization Identifies Clinically Relevant Subgroups of Gastrointestinal Stromal Tumors.

BACKGROUND & AIMS: Gastrointestinal stromal tumor (GIST) is the most common mesenchymal tumor of the gastrointestinal tract, and it has high metastatic and recurrence rates. We aimed to characterize the proteomic features of GIST to understand biological processes and treatment vulnerabilities. METHODS: Quantitative proteomics and phosphoproteomics analyses were performed on 193 patients with GIST to reveal the biological characteristics of GIST. Data-driven hypotheses were tested by performing functional experiments using both GIST cell lines and xenograft mouse models. RESULTS: Proteomic analysis revealed differences in the molecular features of GISTs from different locations or with different histological grades. MAPK7 was identified and functionally proved to be associated with tumor cell proliferation in GIST. Integrative analysis revealed that increased SQSTM1 expression inhibited the patient response to imatinib mesylate. Proteomics subtyping identified 4 clusters of tumors with different clinical and molecular attributes. Functional experiments confirmed the role of SRSF3 in promoting tumor cell proliferation and leading to poor prognosis. CONCLUSIONS: Our study provides a valuable data resource and highlights potential therapeutic approaches for GIST.

Blood molecular markers associated with COVID-19 immunopathology and multi-organ damage.

COVID-19 is characterized by dysregulated immune responses, metabolic dysfunction and adverse effects on the function of multiple organs. To understand host responses to COVID-19 pathophysiology, we combined transcriptomics, proteomics, and metabolomics to identify molecular markers in peripheral blood and plasma samples of 66 COVID-19-infected patients experiencing a range of disease severities and 17 healthy controls. A large number of expressed genes, proteins, metabolites, and extracellular RNAs (exRNAs) exhibit strong associations with various clinical parameters. Multiple sets of tissue-specific proteins and exRNAs varied significantly in both mild and severe patients suggesting a potential impact on tissue function. Chronic activation of neutrophils, IFN-I signaling, and a high level of inflammatory cytokines were observed in patients with severe disease progression. In contrast, COVID-19-infected patients experiencing milder disease symptoms showed robust T-cell responses. Finally, we identified genes, proteins, and exRNAs as potential biomarkers that might assist in predicting the prognosis of SARS-CoV-2 infection. These data refine our understanding of the pathophysiology and clinical progress of COVID-19.

Proteogenomic characterization of cholangiocarcinoma.

BACKGROUND AND AIMS: Cholangiocarcinoma (CCA) is a highly heterogeneous cancer with limited understanding and few effective therapeutic approaches. We aimed at providing a proteogenomic CCA characterization to inform biological processes and treatment vulnerabilities. APPROACH AND RESULTS: Integrative genomic analysis with functional validation uncovered biological perturbations downstream of driver events including DPCR1 , RBM47 mutations, SH3BGRL2 copy number alterations, and FGFR2 fusions in CCA. Proteomic clustering identified three subtypes with distinct clinical outcomes, molecular features, and potential therapeutics. Phosphoproteomics characterized targetable kinases in CCA, suggesting strategies for effective treatment with CDK and MAPK inhibitors. Patients with CCA with HBV infection showed increased antigen processing and presentation (APC) and T cell infiltration, conferring a favorable prognosis compared with those without HBV infection. The characterization of extrahepatic CCA recommended the feasible application of vascular endothelial-derived growth factor inhibitors. Multiomics profiling presented distinctive molecular characteristics of the large bile duct and the small bile duct of intrahepatic CCA. The immune landscape further revealed diverse tumor immune microenvironments, suggesting immune subtypes C1 and C5 might benefit from immune checkpoint therapy. TCN1 was identified as a potential CCA prognostic biomarker, promoting cell growth by enhancing vitamin B12 metabolism. CONCLUSIONS: We characterized the proteogenomic landscape of 217 CCAs with 197 paired normal adjacent tissues and identified their subtypes and potential therapeutic targets. The multiomics analyses with other databases and some functional validations have indicated strategies regarding the clinical, biological, and therapeutic approaches to the management of CCA.

Timing of Cognitive Test Score Decline Prior to Incident Dementia Diagnosis in Blacks and Whites: The Atherosclerosis Risk in Communities Neurocognitive Study.

INTRODUCTION: Commonly occurring dementias include those of Alzheimer's, vascular, and mixtures of these and other pathologies. They are believed to evolve over many years, but that time interval has been difficult to establish. Our objective was to determine how many years in advance of a dementia diagnosis cognitive scores begin to change. METHODS: 14,086 dementia-free ARIC participants underwent a cognitive exam at baseline visit 2 (1990-1992, mean age 57 ± 5.72), and 11,244 at visit 4 (1996-1998), 5,640 at visit 5 (2011-2013), and 3,574 at visit 6 (2016-2017) with surveillance for dementias of all-causes combined. Within 5-year intervals after each visit, we compared performance on the Delayed Word Recall Test (DWRT), the Digit Symbol Substitution Test (DSST), the Word Fluency Test (WFT), and the combined mean of three cognitive tests at baseline in participants who were diagnosed with dementia within each interval versus those who survived the interval without a dementia diagnosis. Z-scores were adjusted for demographics and education in separate regression models for each visit. We plotted adjusted z-score means by time interval following each visit. RESULTS: During follow-up 3,334, 2,821, 1,218, and 329 dementia cases were ascertained after visits 2, 4, 5, and 6, respectively. Adjusted DWRT z-scores were significantly lower 20-25 years before dementia than those who did not experience dementia within 25 years. DSST z-scores were significantly lower at 25-30 years and 3-test combination z-scores were significantly lower as early as 30-31 years before onset. The difference between dementia and non-dementia group in the visit 2 3-test combination z-score was -0.20 at 30-31 years prior to dementia diagnosis. As expected, differences between the dementia and non-dementia groups increased closer to the time of dementia occurrence, up to their widest point at 0-5 years prior to dementia diagnosis. The difference between dementia and non-dementia groups in the visit 2 3-test combination z-score at 0-5 years was -0.90. WFT z-score differences were smaller than for the DSST or DWRT and began later. Patterns were similar in Black and White participants. CONCLUSION: DWRT, DSST, and combined 3-test z-scores were significantly lower more than 20 years prior to diagnosis in the dementia group versus the non-dementia group. Findings contribute to our knowledge of the long prodromal period in Blacks and Whites.

Proteolytic cleavage of amyloid precursor protein by ADAM10 promotes proliferation and migration via activating MAPKs pathway in tongue squamous cell carcinoma in vitro.

APP, well-studied in the development of Alzheimer's disease, has been recently identified as the key gene correlated with TSCC. Here, we investigate the function of APP and its proteolytic cleavage by ADAM10 in the pathogenesis of TSCC. A total of 63 TSCC patients and 30 healthy controls were included and the results of IHC assay showed high expressions of ADAM10 and APP in TSCC tissues compared to paired para-carcinoma tissues. Interestingly, APP expression in TSCC patients was correlated with ADAM10 expression and their combined expression was related to the poor patients' survival. We found that APP was ɑ-cleaved in TSCC cells to form sAPPα, and the serum level of sAPPα but not sAPPß in TSCC patients was higher than healthy controls. Both overexpression with full-length APP and sAPPα promoted TSCC cell proliferation, migration and invasion. Downregulation of APP or ADAM10 by siRNA decreased the generation of sAPPα and inhibited the activity of ERK1/2 and p38 pathways, thereby reducing TSCC cell proliferation, migration and invasion. Treatment with ERK1/2 or p38 agonist or sAPPα overexpression reversed the effects of APP or ADAM10 knockdown. In conclusion, our data demonstrated the pathogenic roles of APP cleaved by ADAM10 to activate ERK1/2 and p38 pathways in TSCC cells. Both high expressions of ADAM10 and APP were related to poor prognosis. Targeting APP cleaved by ADAM10 might be a potential strategy in TSCC treatment.

High Endogenously Synthesized N-3 Polyunsaturated Fatty Acids in Fat-1 Mice Attenuate High-Fat Diet-Induced Insulin Resistance by Inhibiting NLRP3 Inflammasome Activation via Akt/GSK-3ß/TXNIP Pathway.

High-fat (HF) diets and low-grade chronic inflammation contribute to the development of insulin resistance and type 2 diabetes (T2D), whereas n-3 polyunsaturated fatty acids (PUFAs), due to their anti-inflammatory effects, protect against insulin resistance. Interleukin (IL)-1ß is implicated in insulin resistance, yet how n-3 PUFAs modulate IL-1ß secretion and attenuate HF diet-induced insulin resistance remains elusive. In this study, a HF diet activated NLRP3 inflammasome via inducing reactive oxygen species (ROS) generation and promoted IL-1ß production primarily from adipose tissue preadipocytes, but not from adipocytes and induced insulin resistance in wild type (WT) mice. Interestingly, endogenous synthesized n-3 polyunsaturated fatty acids (PUFAs) reversed this process in HF diet-fed fat-1 transgenic mice although the HF diet induced higher weight gain in fat-1 mice, compared with the control diet. Mechanistically, palmitic acid (PA), the main saturated fatty acid in an HF diet inactivated AMPK and led to decreased GSK-3ß phosphorylation, at least partially through reducing Akt activity, which ultimately blocked the Nrf2/Trx1 antioxidant pathway and induced TXNIP cytoplasm translocation and NLRP3 inflammasome activation, whereas docosahexaenoic acid (DHA), the most abundant n-3 PUFA in fat-1 adipose tissue, reversed this process via inducing Akt activation. Our GSK-3ß shRNA knockdown study further revealed that GSK-3ß played a pivot role between the upstream AMPK/Akt pathway and downstream Nrf2/Trx1/TXNIP pathway. Given that NLRP3 inflammasome is implicated in the development of most inflammatory diseases, our results suggest the potential of n-3 PUFAs in the prevention or adjuvant treatment of NLRP3 inflammasome-driven diseases.

Time- and Dose-Resolved Proteome of PM2.5-Exposure-Induced Lung Injury and Repair in Rats.

In recent years, airborne fine particulate matter (PM2.5) is drawing more public attention due to its various physicochemical features and causing pathological harm, as proven by epidemiological and clinical studies. However, the mechanism of PM2.5-exposure-induced lung injury has not been fully characterized. Here, we established a PM2.5-induced rat injury model for both short-term and long-term exposures at different concentrations. We employed the Fast-seq technique to profile 6316 proteins and the catTFRE approach to profile 387 transcription factors (TFs) in the lung tissue. In short-term exposure, we elucidated gradually upregulated proteins enriched in response to oxidative stress, phagosome, and the extracellular matrix (ECM)-receptor interaction pathway. Long-term exposure mainly showed the immune response pathway to be consisting of increased lymphocytes and cytokines. Intriguingly, we found that immune-related proteins were recoverable during short-term exposure. During the process of PM2.5 exposure, upregulated proteins presented dose-dependence in the lung, including stress response at low dose, minor immune response at middle dose, and severe inflammatory response at high dose. This data set provides a rich resource to facilitate the understanding of PM2.5-induced lung damage and repair mechanism.

Polarization Spinodal at Ferroelectric Morphotropic Phase Boundary.

Here, we report a new phenomenon of uniform and continuous transformation of a single polarization domain into alternating nanodomains of two polarization vectors with the same magnitude but different directions at ferroelectric morphotropic phase Boundary (MPB). The transformation is fully reversible and could enhance the piezoelectric coefficient d_{33}. Further free energy calculations illustrate that such a polarization "decomposition" process occurs within the region on the Landau free energy curve with respect to the polarization direction where the second derivative becomes negative, which is similar to spinodal instability in phase transformations such as spinodal ordering or isostructural phase separation (e.g., spinodal decomposition). This "polarization spinodal" uncovers a new mechanism of polarization switching that may account for the ultrahigh ahysterestic piezoelectric strain at the MPB. This work could shed light on the development of phase transition theory and the design of novel ferroelectric memory materials.

A Fully Non-fused Ring Acceptor with Planar Backbone and Near-IR Absorption for High Performance Polymer Solar Cells.

Fused-ring electron acceptors have made significant progress in recent years, while the development of fully non-fused ring acceptors has been unsatisfactory. Here, two fully non-fused ring acceptors, o-4TBC-2F and m-4TBC-2F, were designed and synthesized. By regulating the location of the hexyloxy chains, o-4TBC-2F formed planar backbones, while m-4TBC-2F displayed a twisted backbone. Additionally, the o-4TBC-2F film showed a markedly red-shifted absorption after thermal annealing, which indicated the formation of J-aggregates. For fabrication of organic solar cells (OSCs), PBDB-T was used as a donor and blended with the two acceptors. The o-4TBC-2F-based blend films displayed higher charge mobilities, lower energy loss and a higher power conversion efficiency (PCE). The optimized devices based on o-4TBC-2F gave a PCE of 10.26 %, which was much higher than those based on m-4TBC-2F at 2.63 %, and it is one of the highest reported PCE values for fully non-fused ring electron acceptors.

Proteome-wide Mapping of Endogenous SUMOylation Sites in Mouse Testis.

SUMOylation is a reversible post-translational modification involved in various critical biological processes. To date, there is limited approach for endogenous wild-type SUMO-modified peptides enrichment and SUMOylation sites identification. In this study, we generated a high-affinity SUMO1 antibody to facilitate the enrichment of endogenous SUMO1-modified peptides from Trypsin/Lys-C protease digestion. Following secondary Glu-C protease digestion, we identified 53 high-confidence SUMO1-modified sites from mouse testis by using high-resolution mass spectrometry. Bioinformatics analyses showed that SUMO1-modified proteins were enriched in transcription regulation and DNA repair. Nab1 was validated to be an authentic SUMOylated protein and Lys479 was identified to be the major SUMOylation site. The SUMOylation of Nab1 enhanced its interaction with HDAC2 and maintained its inhibitory effect on EGR1 transcriptional activity. Therefore, we provided a novel approach to investigating endogenous SUMOylation sites in tissue samples.

The role of uric acid in the pathogenesis of diabetic retinopathy based on Notch pathway.

OBJECTIVE: Uric acid has been proposed as an independent risk factor of diabetic retinopathy. Although Notch signaling was reported to be affected in the presence of high concentrations of uric acid or glucose, the underlying mechanisms of hyperuricemia through the Notch signaling pathway to promote the development of diabetic retinopathy remain unknown. METHODS: We incubated human retinal endothelial cells (HRECs) with high glucose, high uric acid and high glucose plus high glucose respectively and evaluated the apoptosis rate in different treated cells by Tunel staining. We induced diabetic model by intraperitoneally streptozotocin. Then healthy rats and diabetic rats were given with adenine and oteracil potassium by gavage. Using automatic biochemical analyzer to detect blood glucose, uric acid, urea nitrogen, creatinine levels, to verify the success of modeling. The expression and mRNA levels of ICAM-1, IL-6, MCP-1, TNF-a, receptors Notch 1, ligands Dll 1, Dll 4, Jagged 1, Jagged 2 were detected by RT-PCR and Western-Blot. Notch1 siRNA was used to interfere Notch signaling pathway, the expression and mRNA levels of ICAM-1, IL-6, MCP-1 and TNF-α was detected by RT-PCR and Western blot respectively. RESULTS: In vitro models, the apoptosis of HRECs cells in high uric acid plus high glucose group was the most significant. In vitro and vivo models, detection of inflammatory cytokines revealed that the expression of inflammatory cytokines increased most significantly in high uric acid plus high glucose group. Notch signaling pathway activity was also increased most significantly in high uric acid plus high glucose group. After Notch 1 siRNA transfection in high glucose and high glucose plus uric acid group, the activity of Notch signaling pathway was successfully down-regulated. We found that the apoptosis of HRECs was significantly decreased in cells transfected with Notch 1 siRNA compared to the blank vector group, and the expression of inflammatory cytokines in cells was also significantly decreased. CONCLUSION: Our study reported that high uric acid can promote the inflammation of the retina and increase the activity of Notch signaling pathway on the basis of high glucose. Hyperuricemia promotes the development of diabetic retinopathy by increasing the activity of Notch signaling pathway. Notch signaling pathway is a potential therapeutic target for diabetic retinopathy.

Lacidipine Amorphous Solid Dispersion Based on Hot Melt Extrusion: Good Miscibility, Enhanced Dissolution, and Favorable Stability.

The present study aimed to increase the in vitro dissolution rate of lacidipine, a poorly water-soluble drug, by formulating amorphous solid dispersions (ASDs) using hot-melt extrusion (HME). Differential scanning calorimetry, powder X-ray diffraction, polarized light microscopy, and Fourier transform infrared were used to characterize the optimal formulations and evaluate the physical stability for the stress test. Film-casting method and hot-stage microscopy were applied to study the miscibility of lacidipine and the drug carriers. In vitro dissolution tests were conducted as the final evaluation index. The optimal formulations were successfully obtained with Soluplus and PVP VA64 at a drug/carrier ratio of 1:10 (w/w), Fourier transform infrared studies revealed the hydrogen bonding between drug and polymers, and in vitro dissolution rates of the optimal formulations were extremely enhanced compared to bulk lacidipine and physical mixtures, similar with that of the commercial tablet. The ASD formulated with Soluplus showed better physical stability than that with PVP VA64. A strong hydrogen bonding and good drug-polymer miscibility were essential to hinder the recrystallization of lacidipine ASDs. In conclusion, the lacidipine ASD formulated with Soluplus showed a significant increase in in vitro dissolution rate and favorable physical stability in the stress test.

Can Patel's τ accurately estimate directionality of connections in brain networks from fMRI?

PURPOSE: Investigating directional interactions between brain regions plays a critical role in fully understanding brain function. Consequently, multiple methods have been developed for noninvasively inferring directional connectivity in human brain networks using functional MRI (fMRI). Recent simulations by Smith et al. showed that Patel's τ, a method based on higher-order statistics, was the best approach for inferring directional interactions from fMRI. Because simulations make restrictive assumptions about reality, we set out to verify this finding using experimental fMRI data obtained from a three-region network in a rat model with electrophysiological validation. METHODS: Previous studies have shown that dynamic causal modeling can correctly estimate the directionality of this three-region network; Granger causality can also work after the deconvolution of the hemodynamic response. Therefore, we set out to test the hypothesis that Patel's τ obtained from either raw or deconvolved fMRI data should correctly estimate the directionality of neuronal influences obtained from intracerebral electroencephalogram in this network. RESULTS: Our results indicate that the accuracy of network directionality estimated using Patel's τ was not better than chance. CONCLUSION: First, our results highlight the necessity of experimental validation of simulation results. Second, it is unclear which brain mechanism is modeled by a directionality inferred from Patel's τ. Third, this study shows that a directional connection ascertained by different methods may mean different things and more experimental studies are needed for investigating the neuronal mechanisms underlying the direction of a connection in the brain ascertained by fMRI using different methods. M Magn Reson Med 78:2003-2010, 2017. © 2017 International Society for Magnetic Resonance in Medicine.

Monte Carlo simulation of magnetic domain structure and magnetic properties near the morphotropic phase boundary.

The morphotropic phase boundary (MPB), which is the boundary separating a tetragonal phase from a rhombohedral phase by varying the composition or mechanical pressure in ferroelectrics, has been studied extensively for decades because it can lead to strong enhancement of piezoelectricity. Recently, a parallel ferromagnetic MPB was experimentally reported in the TbCo2-DyCo2 ferromagnetic system and this discovery proposes a new way to develop potential materials with giant magnetostriction. However, the role of magnetic domain switching and spin reorientation near the MPB region is still unclear. For the first time, we combine micromagnetic theory with Monte Carlo simulation to investigate the evolution of magnetic domain structures and the corresponding magnetization properties near the MPB region. It is demonstrated that the magnetic domain structure and the corresponding magnetization properties are determined by the interplay among anisotropy energy, magnetostatic energy and exchange energy. If the anisotropy energy barrier is large compared with the magnetostatic energy barrier and the exchange energy barrier, the MPB region is a T and R mixed structure and magnetic domain switching is the dominant mechanism. If the anisotropy energy barrier is small, the MPB region will also contain M phases and spin reorientation is the dominant mechanism. Our work could provide a guide for the design of advanced ferromagnetic materials with enhanced magnetostriction.

Genotoxicity testing of sodium formononetin-3'-sulphonate (Sul-F) by assessing bacterial reverse mutation, chromosomal aberrations and micronucleus tests.

As part of a safety evaluation, we evaluated the potential genotoxicity of sodium formononetin-3'-sulphonate (Sul-F) using bacterial reverse mutation assay, chromosomal aberrations detection, and mouse micronucleus test. In bacterial reverse mutation assay using five strains of Salmonella typhimurium (TA97, TA98, TA100, TA102 and TA1535), Sul-F (250, 500, 1000, 2000, 4000 µg/plate) did not increase the number of revertant colonies in any tester strain with or without S9 mix. In a chromosomal assay using Chinese hamster lung fibroblast (CHL) cells, there were no increases in either kind of aberration at any dose of Sul-F (400, 800, and 1600 µg/mL) treatment groups with or without S9 metabolic activation. In an in vivo bone marrow micronucleus test in ICR mice, Sul-F at up to 2000 mg/kg (intravenous injection) showed no significant increases in the incidence of micronucleated polychromatic erythrocytes, and the proportion of immature erythrocytes to total erythrocytes. The results demonstrated that Sul-F does not show mutagenic or genotoxic potential under these test conditions.

Repeated sub-chronic oral toxicity study of xylooligosaccharides (XOS) in dogs.

In this study, Beagle dogs were administered xylooligosaccharide (XOS, CAS # 87099-0) at doses of 0, 1250, 2500, and 5000 mg/kg/day by oral gavage for 26 weeks. A 4-week recovery period was added to observe delayed or reversible toxicity. Measurements included body weight, food consumption, clinical observations, temperature, electrocardiogram (ECG), urinalysis, blood chemistry, hematology, organ weight, gross necropsy, and histopathological examination. Except for transient diarrhea or vomiting, no treatment-related adverse effects were noted. In the mid-dose groups, transitional diarrhea was observed in the initial 1-2 weeks. In the high-dose groups, diarrhea and/or vomiting were observed episodically over the duration of treatment. However, they disappeared after XOS was withdrawn in the recovery period. Although there was a tendency toward less weight gain in the high-dose group animal group, this is typical in animals and humans fed non-digestible carbohydrates. This chronic toxicity study demonstrated that the no observed adverse effect level (NOAEL) of XOS is 2500 mg/kg body weight (BW)/day. Based on body surface area (conversion factor of 0.54 for dogs to human), this corresponds to daily doses of 1350 mg/kg BW or 81-108 g XOS in human adults weighing 60-80 kg.

Acute and a 28-day repeated-dose toxicity study of total flavonoids from Clinopodium chinense (Benth.) O. Ktze in mice and rats.

Clinopodium chinense (Benth.) O. Ktze (Labiatae), known as 'Duanxueliu' in the Chinese Pharmacopoeia, has been widely used as a traditional Chinese medicine for the treatment of hemorrhagic disease. Total flavonoids from Clinopodium chinense (Benth.) O. Ktze (TFCC), the most active ingredient, possess a variety of properties, such as antioxygenation. Until now, evidence-based toxicity data on TFCC has been limited. This study evaluated the acute (in mice and rat) and the 28-day repeated-dose (in rat) toxicity study of TFCC, respectively. In acute study, oral administration of TFCC to rats and mice did not induce toxicity or mortality up to the maximum doses of 4000 and 5000 mg/kg, respectively. In subacute toxicity study, we administered TFCC at daily doses of 70, 210, and 630 mg/kg for 4 consecutive weeks to rats via gavage. We observed no changes in food consumption, water intake, body weight, chemistry and hematological parameters, organ weight, gross pathology or histopathology. No animals from any group died. These findings indicate that TFCC is relatively nontoxic, and provide practical guidance for selecting a safe dose for further investigation of TFCC in animal studies or clinical trials.

Computer-aided classification of mammographic masses using visually sensitive image features.

PURPOSE: To develop a new computer-aided diagnosis (CAD) scheme that computes visually sensitive image features routinely used by radiologists to develop a machine learning classifier and distinguish between the malignant and benign breast masses detected from digital mammograms. METHODS: An image dataset including 301 breast masses was retrospectively selected. From each segmented mass region, we computed image features that mimic five categories of visually sensitive features routinely used by radiologists in reading mammograms. We then selected five optimal features in the five feature categories and applied logistic regression models for classification. A new CAD interface was also designed to show lesion segmentation, computed feature values and classification score. RESULTS: Areas under ROC curves (AUC) were 0.786±0.026 and 0.758±0.027 when to classify mass regions depicting on two view images, respectively. By fusing classification scores computed from two regions, AUC increased to 0.806±0.025. CONCLUSION: This study demonstrated a new approach to develop CAD scheme based on 5 visually sensitive image features. Combining with a "visual aid" interface, CAD results may be much more easily explainable to the observers and increase their confidence to consider CAD generated classification results than using other conventional CAD approaches, which involve many complicated and visually insensitive texture features.

A new approach to develop computer-aided diagnosis scheme of breast mass classification using deep learning technology.

PURPOSE: To develop and test a deep learning based computer-aided diagnosis (CAD) scheme of mammograms for classifying between malignant and benign masses. METHODS: An image dataset involving 560 regions of interest (ROIs) extracted from digital mammograms was used. After down-sampling each ROI from 512×512 to 64×64 pixel size, we applied an 8 layer deep learning network that involves 3 pairs of convolution-max-pooling layers for automatic feature extraction and a multiple layer perceptron (MLP) classifier for feature categorization to process ROIs. The 3 pairs of convolution layers contain 20, 10, and 5 feature maps, respectively. Each convolution layer is connected with a max-pooling layer to improve the feature robustness. The output of the sixth layer is fully connected with a MLP classifier, which is composed of one hidden layer and one logistic regression layer. The network then generates a classification score to predict the likelihood of ROI depicting a malignant mass. A four-fold cross validation method was applied to train and test this deep learning network. RESULTS: The results revealed that this CAD scheme yields an area under the receiver operation characteristic curve (AUC) of 0.696±0.044, 0.802±0.037, 0.836±0.036, and 0.822±0.035 for fold 1 to 4 testing datasets, respectively. The overall AUC of the entire dataset is 0.790±0.019. CONCLUSIONS: This study demonstrates the feasibility of applying a deep learning based CAD scheme to classify between malignant and benign breast masses without a lesion segmentation, image feature computation and selection process.

Transcription Factor Response Elements on Tip: A Sensitive Approach for Large-Scale Endogenous Transcription Factor Quantitative Identification.

The ability to map endogenous transcription factors (TFs) DNA binding activity at the proteome scale will greatly enhance our understanding of various biological processes. Here we report a highly sensitive, rapid, and high-throughput approach, transcription factor response elements on tip-mass spectrometry (TOT-MS), that allows for quantitative measurement of endogenous TFs. A total of 150 TFs from 1 µg of nuclear extracts can be quantified with single shot mass spectrometry detection in 1 h of machine time. Up to 755 TFs, which is comparable to the depth of RNA-seq, were identified by TOT coupled with on-tip small size reverse-phase liquid chromatography. We further demonstrated the capability of TOT-MS by interrogating the dynamic change of TFs in the epidermal growth factor (EGF) signaling pathway. This approach should find broad applications in elucidating the TF landscape from limited amounts of biological materials.