Integrative-omics for discovery of network-level disease biomarkers: a case study in Alzheimer's disease.

A large number of genetic variations have been identified to be associated with Alzheimer's disease (AD) and related quantitative traits. However, majority of existing studies focused on single types of omics data, lacking the power of generating a community including multi-omic markers and their functional connections. Because of this, the immense value of multi-omics data on AD has attracted much attention. Leveraging genomic, transcriptomic and proteomic data, and their backbone network through functional relations, we proposed a modularity-constrained logistic regression model to mine the association between disease status and a group of functionally connected multi-omic features, i.e. single-nucleotide polymorphisms (SNPs), genes and proteins. This new model was applied to the real data collected from the frontal cortex tissue in the Religious Orders Study and Memory and Aging Project cohort. Compared with other state-of-art methods, it provided overall the best prediction performance during cross-validation. This new method helped identify a group of densely connected SNPs, genes and proteins predictive of AD status. These SNPs are mostly expression quantitative trait loci in the frontal region. Brain-wide gene expression profile of these genes and proteins were highly correlated with the brain activation map of 'vision', a brain function partly controlled by frontal cortex. These genes and proteins were also found to be associated with the amyloid deposition, cortical volume and average thickness of frontal regions. Taken together, these results suggested a potential pathway underlying the development of AD from SNPs to gene expression, protein expression and ultimately brain functional and structural changes.

A randomized controlled trial of WATAAP to promote physical activity in colorectal and endometrial cancer survivors.

OBJECTIVE: The objective of this study was to ascertain whether wearable technology coupled with action planning was effective in increasing physical activity (PA) in colorectal and endometrial cancer survivors at cardiovascular risk. METHODS: Sixty-eight survivors who had cardiovascular risk factors and were insufficiently active were randomized to intervention and control arms. Intervention participants were given a wearable tracker for 12 weeks, two group sessions, and a support phone call. Participants in the control arm received print materials describing PA guidelines. Assessments at baseline and 12 weeks measured triaxial and uniaxial estimates of moderate-vigorous physical activity (MVPA), sedentary behaviour, blood pressure, and body mass index (BMI). RESULTS: The intervention group significantly increased MVPA by 45 min/wk compared with a reduction of 21 min/wk in the control group. Group by time interactions were significant for minutes of MVPA (F1,126  = 5.14, P = 0.025). For those with diastolic hypertension, there was a significant group by time interaction (F1,66  = 4.89, P = 0.031) with a net reduction of 9.89 mm Hg in the intervention group. CONCLUSIONS: Significant improvements in MVPA were observed following the intervention. The results display promise for the use of pragmatic, low-intensity interventions using wearable technology.

Population-based screening for Lynch syndrome in Western Australia.

We showed earlier that routine screening for microsatellite instability (MSI) and loss of mismatch repair (MMR) protein expression in colorectal cancer (CRC) led to the identification of previously unrecognized cases of Lynch syndrome (LS). We report here the results of screening for LS in Western Australia (WA) during 1994-2012. Immunohistochemistry (IHC) for loss of MMR protein expression was performed in routine pathology laboratories, while MSI was detected in a reference molecular pathology laboratory. Information on germline mutations in MMR genes was obtained from the state's single familial cancer registry. Prior to the introduction of routine laboratory-based screening, an average of 2-3 cases of LS were diagnosed each year amongst WA CRC patients. Following the implementation of IHC and/or MSI screening for all younger (<60 years) CRC patients, this has increased to an average of 8 LS cases diagnosed annually. Based on our experience in WA, we propose three key elements for successful population-based screening of LS. First, for all younger CRC patients, reflex IHC testing should be carried out in accredited pathology services with ongoing quality control. Second, a state- or region-wide reference laboratory for MSI testing should be established to confirm abnormal or suspicious IHC test results and to exclude sporadic cases by carrying out BRAF mutation or MLH1 methylation testing. Finally, a state or regional LS coordinator is essential to ensure that all appropriate cases identified by laboratory testing are referred to and attend a Familial Cancer Clinic for follow-up and germline testing.

A novel suspension formulation enhances intestinal absorption of macromolecules via transient and reversible transport mechanisms.

PURPOSE: Medium chain fatty acid salts promote absorption by increasing paracellular permeability of the intestinal epithelium. Novel oily suspension (OS) formulation disperses a powder containing sodium caprylate and macromolecules such as octreotide or fluorescent dextran (FD). Formulation safety, macromolecule absorption and pharmacokinetic (PK)/pharmacodynamic (PD) were evaluated. METHODS: Octreotide/OS toxicity was evaluated in monkeys following 9 months of daily oral enteric-coated capsule administration. The OS permeation effect was also assessed in rats, using FD/OS and octreotide/OS preparations. Octreotide/OS effects on circulating growth hormone (GH) levels were also measured. RESULTS: Safety assessment of octreotide/OS in monkeys after 9 months showed minor drug-related findings, comparable to the injectable octreotide. Octreotide exposure levels were similar across the treatment periods. In rats, OS facilitated FD permeation up to 70 kDa in a reversible, spatial and dose-dependent manner, independent of the intestinal dosing site. Following OS administration, the staining pattern of the tight-junction protein, ZO-1, changed transiently, and a paracellular penetration marker, LC-biotin, permeated between adjacent epithelial cells. Enteral octreotide/OS absorption was dose-dependent and suppressed rat GH levels. CONCLUSIONS: Oral octreotide/OS dosing was shown to be safe in monkeys. OS enhances intestinal absorption of active octreotide, likely by transient alteration of the tight junction protein complex.

Deep Trans-Omic Network Fusion for Molecular Mechanism of Alzheimer's Disease.

Background: There are various molecular hypotheses regarding Alzheimer's disease (AD) like amyloid deposition, tau propagation, neuroinflammation, and synaptic dysfunction. However, detailed molecular mechanism underlying AD remains elusive. In addition, genetic contribution of these molecular hypothesis is not yet established despite the high heritability of AD. Objective: The study aims to enable the discovery of functionally connected multi-omic features through novel integration of multi-omic data and prior functional interactions. Methods: We propose a new deep learning model MoFNet with improved interpretability to investigate the AD molecular mechanism and its upstream genetic contributors. MoFNet integrates multi-omic data with prior functional interactions between SNPs, genes, and proteins, and for the first time models the dynamic information flow from DNA to RNA and proteins. Results: When evaluated using the ROS/MAP cohort, MoFNet outperformed other competing methods in prediction performance. It identified SNPs, genes, and proteins with significantly more prior functional interactions, resulting in three multi-omic subnetworks. SNP-gene pairs identified by MoFNet were mostly eQTLs specific to frontal cortex tissue where gene/protein data was collected. These molecular subnetworks are enriched in innate immune system, clearance of misfolded proteins, and neurotransmitter release respectively. We validated most findings in an independent dataset. One multi-omic subnetwork consists exclusively of core members of SNARE complex, a key mediator of synaptic vesicle fusion and neurotransmitter transportation. Conclusions: Our results suggest that MoFNet is effective in improving classification accuracy and in identifying multi-omic markers for AD with improved interpretability. Multi-omic subnetworks identified by MoFNet provided insights of AD molecular mechanism with improved details.

NISNet3D: three-dimensional nuclear synthesis and instance segmentation for fluorescence microscopy images.

The primary step in tissue cytometry is the automated distinction of individual cells (segmentation). Since cell borders are seldom labeled, cells are generally segmented by their nuclei. While tools have been developed for segmenting nuclei in two dimensions, segmentation of nuclei in three-dimensional volumes remains a challenging task. The lack of effective methods for three-dimensional segmentation represents a bottleneck in the realization of the potential of tissue cytometry, particularly as methods of tissue clearing present the opportunity to characterize entire organs. Methods based on deep learning have shown enormous promise, but their implementation is hampered by the need for large amounts of manually annotated training data. In this paper, we describe 3D Nuclei Instance Segmentation Network (NISNet3D) that directly segments 3D volumes through the use of a modified 3D U-Net, 3D marker-controlled watershed transform, and a nuclei instance segmentation system for separating touching nuclei. NISNet3D is unique in that it provides accurate segmentation of even challenging image volumes using a network trained on large amounts of synthetic nuclei derived from relatively few annotated volumes, or on synthetic data obtained without annotated volumes. We present a quantitative comparison of results obtained from NISNet3D with results obtained from a variety of existing nuclei segmentation techniques. We also examine the performance of the methods when no ground truth is available and only synthetic volumes were used for training.

Generalized Matrix Local Low Rank Representation by Random Projection and Submatrix Propagation.

Matrix low rank approximation is an effective method to reduce or eliminate the statistical redundancy of its components. Compared with the traditional global low rank methods such as singular value decomposition (SVD), local low rank approximation methods are more advantageous to uncover interpretable data structures when clear duality exists between the rows and columns of the matrix. Local low rank approximation is equivalent to low rank submatrix detection. Unfortunately, existing local low rank approximation methods can detect only submatrices of specific mean structure, which may miss a substantial amount of true and interesting patterns. In this work, we develop a novel matrix computational framework called RPSP (Random Probing based submatrix Propagation) that provides an effective solution for the general matrix local low rank representation problem. RPSP detects local low rank patterns that grow from small submatrices of low rank property, which are determined by a random projection approach. RPSP is supported by theories of random projection. Experiments on synthetic data demonstrate that RPSP outperforms all state-of-the-art methods, with the capacity to robustly and correctly identify the low rank matrices when the pattern has a similar mean as the background, background noise is heteroscedastic and multiple patterns present in the data. On real-world datasets, RPSP also demonstrates its effectiveness in identifying interpretable local low rank matrices.

Artificial intelligence reveals features associated with breast cancer neoadjuvant chemotherapy responses from multi-stain histopathologic images.

Advances in computational algorithms and tools have made the prediction of cancer patient outcomes using computational pathology feasible. However, predicting clinical outcomes from pre-treatment histopathologic images remains a challenging task, limited by the poor understanding of tumor immune micro-environments. In this study, an automatic, accurate, comprehensive, interpretable, and reproducible whole slide image (WSI) feature extraction pipeline known as, IMage-based Pathological REgistration and Segmentation Statistics (IMPRESS), is described. We used both H&E and multiplex IHC (PD-L1, CD8+, and CD163+) images, investigated whether artificial intelligence (AI)-based algorithms using automatic feature extraction methods can predict neoadjuvant chemotherapy (NAC) outcomes in HER2-positive (HER2+) and triple-negative breast cancer (TNBC) patients. Features are derived from tumor immune micro-environment and clinical data and used to train machine learning models to accurately predict the response to NAC in breast cancer patients (HER2+ AUC = 0.8975; TNBC AUC = 0.7674). The results demonstrate that this method outperforms the results trained from features that were manually generated by pathologists. The developed image features and algorithms were further externally validated by independent cohorts, yielding encouraging results, especially for the HER2+ subtype.

FOXP3+ cell density in lymphoid follicles from histologically normal mucosa is a strong prognostic factor in early stage colon cancer.

There are few clearly established prognostic factors available to guide the use of adjuvant chemotherapy in early stage colon cancer patients. Some of the most promising candidates include the invasion of extramural blood vessels by tumour cells and the densities of FOXP3+ T regulatory cells (Tregs) in tumour and adjacent normal colonic mucosal tissue. The aim of our study was to evaluate the prognostic significance of these markers in AJCC stage II colon cancer, with particular reference to lymphoid follicles in the mucosa. Histopathological review for the presence of vascular and serosal invasion was conducted on a series of 165 stage II colon cancers treated by surgery alone. Immunohistochemical staining for FOXP3 was performed on tumour tissue and histologically normal colonic mucosa from the surgical margin. Image analysis software was used to evaluate the density of FOXP3+ cells in the tumour core, invading margin and lymphoid follicles from the colonic mucosa. For survival analysis, cases were classified into high- or low-density of FOXP3+ cells according to the median value. The mean density of FOXP3+ Tregs in lymphoid follicles was twofold and fivefold higher than in the invading margin and tumour core, respectively. Multivariate analysis identified extramural vascular invasion (HR, 2.47; 95% CI: 1.00-6.07; P = 0.05) and high FOXP3+ cell density in lymphoid follicles (HR, 4.22; 95% CI: 1.49-11.91; P = 0.007) as independent factors for worse survival, whereas a high frequency of lymphoid follicles in histologically normal colonic mucosa was associated with better survival (HR, 0.31; 95% CI: 0.12-0.79; P = 0.014). Our data suggest that host factors related to the immune system have major prognostic significance in early stage colon cancer. The density of FOXP3+ cells within lymphoid follicles and the frequency of these structures in normal colonic mucosa represent novel and independent prognostic factors.

Two-photon imaging within the murine thorax without respiratory and cardiac motion artifact.

Intravital microscopy has been recognized for its ability to make physiological measurements at cellular and subcellular levels while maintaining the complex natural microenvironment. Two-photon microscopy (TPM), using longer wavelengths than single-photon excitation, has extended intravital imaging deeper into tissues, with minimal phototoxicity. However, due to a relatively slow acquisition rate, TPM is especially sensitive to motion artifact, which presents a challenge when imaging tissues subject to respiratory and cardiac movement. Thoracoabdominal organs that cannot be exteriorized or immobilized during TPM have generally required the use of isolated, pump-perfused preparations. However, this approach entails significant alteration of normal physiology, such as a lack of neural inputs, increased vascular resistance, and leukocyte activation. We adapted techniques of intravital microscopy that permitted TPM of organs maintained within the thoracoabdominal cavity of living, breathing rats or mice. We obtained extended intravital TPM imaging of the intact lung, arguably the organ most susceptible to both respiratory and cardiac motion. Intravital TPM detected the development of lung microvascular endothelial activation manifested as increased leukocyte adhesion and plasma extravasation in response to oxidative stress inducers PMA or soluble cigarette smoke extract. The pulmonary microvasculature and alveoli in the intact animal were imaged with comparable detail and fidelity to those in pump-perfused animals, opening the possibility for TPM of other thoracoabdominal organs under physiological and pathophysiological conditions.

Antibiotic practice patterns with procalcitonin levels in patients with acute lower respiratory tract infection.

OBJECTIVES: Procalcitonin (PCT) testing is FDA approved to guide antibiotic therapy in patients with lower respiratory tract infection (LRTI). However, its utilization and impact on real-world antibiotic prescribing behavior are unknown. We investigated the rate of PCT testing to evaluate an association between initial PCT level and antibiotic prescription patterns for patients with suspected LRTI within a large integrated health system. STUDY DESIGN: Retrospective cohort study. METHODS: A retrospective cohort study (January 1, 2016, through December 31, 2017) was performed in patients 18 years and older who were hospitalized with LRTI and had a PCT measurement. Antibiotic changes were noted before and 36 hours after initial PCT results. Antibiotic concordance was determined using a PCT cutoff value of 0.25 mcg/L. Concordance was defined as (1) patients received antibiotics after a PCT of at least 0.25 mcg/L resulted or (2) antibiotics were withheld after a PCT less than 0.25 mcg/L resulted. RESULTS: PCT testing occurred in 18% of hospitalized patients with LRTI. Among 1606 patients, antibiotic concordance with PCT results was 55%. Among the discordant population, 77% of patients received antibiotics in the setting of a low PCT level compared with 23% who did not receive antibiotics at a high PCT level. There were no statistical differences between LRTI types between patients with PCT-discordant and PCT-concordant care. CONCLUSIONS: Within a real-world environment of patients hospitalized with LRTI, PCT testing was low and the PCT levels did not appear to influence antibiotic prescribing behavior. Our findings suggest that clinicians continue to prioritize clinical judgment over initial PCT levels when prescribing antibiotics for suspected LRTIs.

Low expression of Granzyme B in colorectal cancer is associated with signs of early metastastic invasion.

AIMS: Tumour-infiltrating forkhead box P3 (FoxP3+ ) regulatory T cells (T(regs) ) have stronger prognostic significance than cytotoxic CD8+ T cells in colorectal cancer (CRC). Because there is evidence that some tumour-infiltrating CD8+ T cells may be inactive, the present study aimed to investigate the prognostic significance of Granzyme B, one of the major effector molecules of T cells. METHODS AND RESULTS: A tissue microarray containing 963 CRCs was stained immunohistochemically for Granzyme B and the level of expression quantified by digital image analysis. Granzyme B expression was higher in tumours with microsatellite instability (P < 0.0001), a dense lymphocytic infiltrate (P < 0.0001) and location in the proximal colon (P = 0.009), but lower in tumours with vascular invasion (P = 0.007), perineural invasion (P =0.041) and positive nodal status (P < 0.001). Elevated expression of Granzyme B was associated with improved survival on univariate analysis (hazard ratio = 0.65; 95% confidence interval 0.51-0.84; P = 0.001), but not in a multivariate model that included stage, vascular invasion and FoxP3+ T(reg) cell density. CONCLUSIONS: Low expression of Granzyme B was associated with early signs of metastasis in CRC. The stronger prognostic significance of FoxP3+ T(regs) is in keeping with animal models that suggest these cells act as gatekeepers for the release of Granzyme B from CD8+ T cells.

Western Australian Peritonectomy Service: a review of the first 6 years of operation of a low-volume peritonectomy centre.

BACKGROUND: Cytoreductive surgery (CRS) with hyperthermic intraperitoneal early chemotherapy has gained acceptance as the standard of care for peritoneal surface malignancy over the past decade. This study reports the results of the first 6 years of activity of the Western Australian Peritonectomy Service to compare outcomes of a low-volume centre with world standards. METHODS: An audit of all patients who received CRS ± hyperthermic intraperitoneal early chemotherapy treatment at the Western Australian Peritonectomy Service in its first 6 years of operation was performed and results were compared to the recent world literature. RESULTS: A total of 130 patients were treated with 150 CRS procedures, including 50 pseudomyxoma peritonei (PMP), 53 colorectal cancers (CRCs), 27 appendix adenocarcinomas (AAs) and 20 other cancers. The median operating time was 12 h with median length of hospital stay of 15 days. Perioperative mortality was 0.67% and Clavien-Dindo III/IV morbidity was 20%. The mean packed red cell transfusion requirement was 0.34 units/case. Post-discharge readmission rate was 25% and the return to theatre rate was 13%. The 4-year overall survival rates for PMP, CRC and AA were 97%, 49% and 81%, respectively. The 3-year disease-free survival rates for PMP, CRC and AA were 74%, 26% and 36%, respectively. CONCLUSION: A CRS centre averaging 25 cases per year can achieve strong outcomes in line with high-volume world centres.

Eletromagnetic Detection of Mild Brain Injury: A Novel Imaging Approach to Post Concussive Syndrome.

INTRODUCTION: Mild traumatic brain injury (mTBI) is a common injury, with nearly 3 - 4 million cases annually in the United States alone. Neuroimaging in patients with mTBI provides little benefit, and is usually not indicated as the diagnosis is primarily clinical. It is theorized that microvascular trauma to the brain may be present in mTBI, that may not be captured by routine MRI and CT scans. Electromagnetic (EM) waves may provide a more sensitive medical imaging modality to provide objective data in the diagnosis of mTBI. METHODS: COMSOL simulation software was utilized to mimic the anatomy of the human skull including skin, cranium, cerebrospinal fluid (CSF), gray-matter tissue of the brain, and microvasculature within the neural tissue. The effects of penetrating EM waves were simulated using the finite element analysis software and results were generated to identify feasibility and efficacy. Frequency ranges from 7 GHz to 15 GHz were considered, with 0.6 and 1 W power applied. RESULTS: Variations between the differing frequency levels generated different energy levels within the neural tissue-particularly when comparing normal microvasculature versus hemorrhage from microvasculature. This difference within the neural tissue was subsequently identified, via simulation, serving as a potential imaging modality for future work. CONCLUSION: The use of electromagnetic imaging of the brain after concussive events may play a role in future mTBI diagnosis. Utilizing the proper depth frequency and wavelength, neural tissue and microvascular trauma may be identified utilizing finite element analysis.

TSUNAMI: Translational Bioinformatics Tool Suite for Network Analysis and Mining.

Gene co-expression network (GCN) mining identifies gene modules with highly correlated expression profiles across samples/conditions. It enables researchers to discover latent gene/molecule interactions, identify novel gene functions, and extract molecular features from certain disease/condition groups, thus helping to identify disease biomarkers. However, there lacks an easy-to-use tool package for users to mine GCN modules that are relatively small in size with tightly connected genes that can be convenient for downstream gene set enrichment analysis, as well as modules that may share common members. To address this need, we developed an online GCN mining tool package: TSUNAMI (Tools SUite for Network Analysis and MIning). TSUNAMI incorporates our state-of-the-art lmQCM algorithm to mine GCN modules for both public and user-input data (microarray, RNA-seq, or any other numerical omics data), and then performs downstream gene set enrichment analysis for the identified modules. It has several features and advantages: 1) a user-friendly interface and real-time co-expression network mining through a web server; 2) direct access and search of NCBI Gene Expression Omnibus (GEO) and The Cancer Genome Atlas (TCGA) databases, as well as user-input gene expression matrices for GCN module mining; 3) multiple co-expression analysis tools to choose from, all of which are highly flexible in regards to parameter selection options; 4) identified GCN modules are summarized to eigengenes, which are convenient for users to check their correlation with other clinical traits; 5) integrated downstream Enrichr enrichment analysis and links to other gene set enrichment tools; and 6) visualization of gene loci by Circos plot in any step of the process. The web service is freely accessible through URL: https://biolearns.medicine.iu.edu/. Source code is available at https://github.com/huangzhii/TSUNAMI/.

The wearable activity technology and action-planning trial in cancer survivors: Physical activity maintenance post-intervention.

OBJECTIVES: The study objective was to assess whether moderate-to-vigorous intensity physical activity (MVPA) change in cancer survivors (n = 68, mean age = 64 years) was maintained 12-weeks following the Wearable Activity Technology and Action Planning (WATAAP) intervention. Secondary aims were to assess the effects of the intervention on blood pressure (BP) and body mass index (BMI), and to explore group differences between baseline and 24-weeks. DESIGN: Randomized controlled trial. METHODS: MVPA and sedentary behaviour were assessed using an accelerometer at baseline, the end of the intervention (12-weeks), and at 24-weeks. Generalised linear mixed models with random effects were used to examine between-group and within-group changes in MVPA, sedentary behaviour, BP and BMI. RESULTS: MVPA was significantly higher in the intervention group compared with the control group at 24-weeks following adjustment for known confounders (141.4 min/wk. (95% CI = 9.1 to 273.8), p = 0.036). At 24-weeks participants in the intervention group had maintained their increased levels of MVPA (change from 12-weeks = 8.8 min/wk.; 95% CI = -43 to 61; p = 0.74). The reduction in MVPA in the control group over the first 12-weeks was also maintained at 24-weeks (5.4 min/wk.; 95% CI = -3.6 to 4.6; p = 0.80). Secondary outcomes did not differ between groups at 24-weeks. CONCLUSIONS: Our results suggest distance-based interventions using wearable technology produce increases in MVPA that endure at least 12-weeks after the intervention is completed.

Identification of functionally connected multi-omic biomarkers for Alzheimer's disease using modularity-constrained Lasso.

Large-scale genome wide association studies (GWASs) have led to discovery of many genetic risk factors in Alzheimer's disease (AD), such as APOE, TOMM40 and CLU. Despite the significant progress, it remains a major challenge to functionally validate these genetic findings and translate them into targetable mechanisms. Integration of multiple types of molecular data is increasingly used to address this problem. In this paper, we proposed a modularity-constrained Lasso model to jointly analyze the genotype, gene expression and protein expression data for discovery of functionally connected multi-omic biomarkers in AD. With a prior network capturing the functional relationship between SNPs, genes and proteins, the newly introduced penalty term maximizes the global modularity of the subnetwork involving selected markers and encourages the selection of multi-omic markers with dense functional connectivity, instead of individual markers. We applied this new model to the real data collected in the ROS/MAP cohort where the cognitive performance was used as disease quantitative trait. A functionally connected subnetwork involving 276 multi-omic biomarkers, including SNPs, genes and proteins, were identified to bear predictive power. Within this subnetwork, multiple trans-omic paths from SNPs to genes and then proteins were observed. This suggests that cognitive performance deterioration in AD patients can be potentially a result of genetic variations due to their cascade effect on the downstream transcriptome and proteome level.

Sphere estimation network: three-dimensional nuclei detection of fluorescence microscopy images.

Purpose: Fluorescence microscopy visualizes three-dimensional subcellular structures in tissue with two-photon microscopy achieving deeper penetration into tissue. Nuclei detection, which is essential for analyzing tissue for clinical and research purposes, remains a challenging problem due to the spatial variability of nuclei. Recent advancements in deep learning techniques have enabled the analysis of fluorescence microscopy data to localize and segment nuclei. However, these localization or segmentation techniques would require additional steps to extract characteristics of nuclei. We develop a 3D convolutional neural network, called Sphere Estimation Network (SphEsNet), to extract characteristics of nuclei without any postprocessing steps. Approach: To simultaneously estimate the center locations of nuclei and their sizes, SphEsNet is composed of two branches to localize nuclei center coordinates and to estimate their radii. Synthetic microscopy volumes automatically generated using a spatially constrained cycle-consistent adversarial network are used for training the network because manually generating 3D real ground truth volumes would be extremely tedious. Results: Three SphEsNet models based on the size of nuclei were trained and tested on five real fluorescence microscopy data sets from rat kidney and mouse intestine. Our method can successfully detect nuclei in multiple locations with various sizes. In addition, our method was compared with other techniques and outperformed them based on object-level precision, recall, and F 1 score. Our model achieved 89.90% for F 1 score. Conclusions: SphEsNet can simultaneously localize nuclei and estimate their size without additional steps. SphEsNet can be potentially used to extract more information from nuclei in fluorescence microscopy images.

Differential co-expression analysis reveals early stage transcriptomic decoupling in alzheimer's disease.

BACKGROUND: Alzheimer's disease (AD) is one of the leading causes of death in the US and there is no validated drugs to stop, slow or prevent AD. Despite tremendous effort on biomarker discovery, existing findings are mostly individual biomarkers and provide limited insights into the transcriptomic decoupling underlying AD. We propose to explore the gene co-expression patterns in multiple AD stages, including cognitively normal (CN), early mild cognitive impairment (EMCI), late MCI and AD. METHODS: We modified traiditonal joint graphical lasso to model our asusmption that the co-expression networks in consecutive disease stages are largely similar with critical differences. In addition, we performed subsequent network comparison analysis for identification of stage specific transcriptomic decoupling. We focused our analysis on top AD-enriched pathways. RESULTS: We observed that 419 edges in CN, 420 edges in EMCI, 381 edges in LMCI and 250 edges in AD were frequently estimated with non zero weights. With modified JGL, the weight of all estimated edges in CN, EMCI and LMCI are zero. In AD group, 299 edges were occasionally estimated to be nonzero and the average correlation between genes was 0.0023. For co-expression change during AD progression, there are 66 pairs of genes that demonstrated a continuously decreasing or increasing co-expression from CN to EMCI, LMCI and AD.The network level clustering coefficient remains stable from CN to LMCI and then decreases significantly when progressing to AD. When evaluating edge level differences, we identified eight gene modules with continuously decreasing or increasing co-expression patterns during AD progression. Five of them shows significant changes from CN to EMCI and thus have the potential to serve system biomarkers for early screening of AD. CONCLUSION: We employed a modified joint graphical lasso for estimation of co-expression networks for multiple stages of AD. Comparing with graphical lasso, our modified joint graphical lasso model accounts for the similarity in consecutive disease stages. Our results on real data set revealed five gene clusters with obvious co-expression pattern change from CN to EMCI, which could be used as potential system-level biomarkers for early screening of AD.