Search

Traumatic brain injury (TBI) is a major cause of motor and cognitive impairment in young adults. It is associated with high mortality rates and very few effective treatment options. Bisperoxovanadium (pyridine-2-carboxyl) [bpV(pic)] is an commercially available inhibitor of Phosphatase and tensin homolog (PTEN). Previous studies have shown that bpV(pic) has protective effects in central nervous system. However, the role of bpV(pic) in TBI is unclear. In this study we aimed to investigate the neuroprotective role of bpV(pic) in rat TBI model. We found that injection of bpV(pic) significantly reduces brain edema and neurological dysfunction after TBI and this is mediated by AKT pathway. TBI is known to promote the M1 pro-inflammatory phenotype of microglial polarization and this effect is inhibited by bpV(pic) treatment which, instead promotes M2 microglial polarization in vivo and in vitro. We also found evidence of bpV(pic)-regulated neuroinflammation mediated by AKT activation and NF-κB p65 inhibition. BpV(pic) treatment also suppressed microglia in the peri-TBI region. MCP-1 is known to recruit monocytes and macrophages to promote inflammation, we show that bpV(pic) can inhibit TBI-induced up-regulation of MCP-1 via the AKT/NF-κB p65 signaling pathway. Taken together, our findings demonstrate that bpV(pic) plays a neuroprotective role in rat TBI, which may be achieved by inhibiting M1 microglia polarization and MCP-1 expression by modulating AKT/NF-κB p65 signaling pathway.

Glycine is a simple nonessential amino acid known to have neuroprotective properties. Treatment with glycine results in reduced infarct volume of the brain, neurologic function scores, and neuronal and microglial death in ischemic stroke injury. Neuroinflammation has been considered a major contributor to cerebral ischemia–induced brain damage. However, the role of glycine in neuroinflammation following ischemic stroke is unclear. The present study aimed to determine whether neuroinflammation is involved in the neuroprotective effects of glycine in cerebral ischemia injury. Ischemic stroke promotes M1 microglial polarization. Interestingly, we found that the injection of glycine in rats after injury can inhibit ischemia-induced inflammation and promote M2 microglial polarization in vivo (Sprague–Dawley rats) and in vitro (cortical microglia and BV-2 cells). We show that glycine suppresses Hif-1α by inhibiting the upregulation of NF-κB p65 after ischemia-reperfusion injury, resulting in the inhibition of proinflammatory activity. The activation of AKT mediates the inhibition of NF-κB p65/Hif-1α signaling by glycine. Moreover, we confirm that glycine-regulated AKT activation is mediated by the inhibition of PTEN in a PTEN depletion cell line, U251 cells. Glycine modulates microglial polarization after ischemic stroke, which indirectly inhibits ischemia-induced neuronal death and functional recovery. Taken together, our findings provide a new understanding of glycine in neuroprotection by inhibiting M1 microglial polarization and promoting anti-inflammation by suppressing NF-κB p65/Hif-1α signaling.

Breast cancer stem cells (BCSCs) contribute to intra-tumoral heterogeneity and therapeutic resistance. However, the binary concept of universal BCSCs co-existing with bulk tumor cells is over-simplified. Through single-cell RNA-sequencing, we found that Neu, PyMT and BRCA1-null mammary tumors each corresponded to a spectrum of minimally overlapping cell differentiation states without a universal BCSC population. Instead, our analyses revealed that these tumors contained distinct lineage-specific tumor propagating cells (TPCs) and this is reflective of the self-sustaining capabilities of lineage-specific stem/progenitor cells in the mammary epithelial hierarchy. By understanding the respective tumor hierarchies, we were able to identify CD14 as a TPC marker in the Neu tumor. Additionally, single-cell breast cancer subtype stratification revealed the co-existence of multiple breast cancer subtypes within tumors. Collectively, our findings emphasize the need to account for lineage-specific TPCs and the hierarchical composition within breast tumors, as these heterogenous sub-populations can have differential therapeutic susceptibilities.

Ongoing biomarker development programs have been designed to identify serologic or imaging signatures of clinico-pathologic entities, assuming distinct biological boundaries between them. Identified putative biomarkers have exhibited large variability and inconsistency between cohorts, and remain inadequate for selecting suitable recipients for potential disease-modifying interventions. We launched the Cincinnati Cohort Biomarker Program (CCBP) as a population-based, phenotype-agnostic longitudinal study. While patients affected by a wide range of neurodegenerative disorders will be deeply phenotyped using clinical, imaging, and mobile health technologies, analyses will not be anchored on phenotypic clusters but on bioassays of to-be-repurposed medications as well as on genomics, transcriptomics, proteomics, metabolomics, epigenomics, microbiomics, and pharmacogenomics analyses blinded to phenotypic data. Unique features of this cohort study include (1) a reverse biology-to-phenotype direction of biomarker development in which clinical, imaging, and mobile health technologies are subordinate to biological signals of interest; (2) hypothesis free, causally- and data driven-based analyses; (3) inclusive recruitment of patients with neurodegenerative disorders beyond clinical criteria-meeting patients with Parkinson's and Alzheimer's diseases, and (4) a large number of longitudinally followed participants. The parallel development of serum bioassays will be aimed at linking biologically suitable subjects to already available drugs with repurposing potential in future proof-of-concept adaptive clinical trials. Although many challenges are anticipated, including the unclear pathogenic relevance of identifiable biological signals and the possibility that some signals of importance may not yet be measurable with current technologies, this cohort study abandons the anchoring role of clinico-pathologic criteria in favor of biomarker-driven disease subtyping to facilitate future biosubtype-specific disease-modifying therapeutic efforts.

Background Traditionally, the impact of lipoproteins on vascular disease has been evaluated in light of their quantity, that is, cholesterol content, in plasma. However, recent studies of high-density lipoproteins (HDLs) have focused on functionality with regard to atheroprotection. For example, bioassays have emerged to assess the ability of HDL, in its near native plasma environment, to promote cholesterol removal (efflux) from cells. As a result, attention has focused on developing plasma-based assays for other putative HDL protective functions including protecting low-density lipoproteins (LDLs) from oxidative damage. Objective To determine the feasibility of such an assay in a complex sample such as plasma, we evaluated the contribution of HDL vs other plasma factors in preventing LDL oxidation. Methods We separated normolipidemic human plasma by gel filtration chromatography and assessed each fraction for its ability to prevent LDL modification by water soluble radical and copper-initiated oxidation mechanisms. Results Using proteomics and selective precipitation methods, we identified major antioxidative contributions for fibrinogen, immunoglobulin G, albumin, and small soluble molecules like uric acid and ascorbate, with albumin being especially dominant in copper-initiated mechanisms. HDL particles were minor contributors (∼1%–2%) to the antioxidant capacity of plasma, irrespective of oxidation mechanism. Conclusions Given the overwhelming background of antioxidant capacity inherent to highly abundant plasma proteins, specific bioassays of HDL antioxidative function will likely require its complete separation from plasma.

The trends in cochlear implantation candidacy and benefit have changed rapidly in the last two decades. It is now widely accepted that early implantation leads to better postimplant outcomes. Although some generalizations can be made about postimplant auditory and language performance, neural mechanisms need to be studied to predict individual prognosis.The aim of this study was to use functional magnetic resonance imaging (fMRI) to identify preimplant neuroimaging biomarkers that predict children’s postimplant auditory and language outcomes as measured by parental observation/reports.This is a pre–post correlational measures study.Twelve possible cochlear implant candidates with bilateral severe to profound hearing loss were recruited via referrals for a clinical magnetic resonance imaging to ensure structural integrity of the auditory nerve for implantation.Participants underwent cochlear implantation at a mean age of 19.4 mo. All children used the advanced combination encoder strategy (ACE, Cochlear Corporation™, Nucleus® Freedom cochlear implants). Three participants received an implant in the right ear; one in the left ear whereas eight participants received bilateral implants. Participants’ preimplant neuronal activation in response to two auditory stimuli was studied using an event-related fMRI method.Blood oxygen level dependent contrast maps were calculated for speech and noise stimuli. The general linear model was used to create z-maps. The Auditory Skills Checklist (ASC) and the SKI-HI Language Development Scale (SKI-HI LDS) were administered to the parents 2 yr after implantation. A nonparametric correlation analysis was implemented between preimplant fMRI activation and postimplant auditory and language outcomes based on ASC and SKI-HI LDS. Statistical Parametric Mapping software was used to create regression maps between fMRI activation and scores on the aforementioned tests. Regression maps were overlaid on the Imaging Research Center infant template and visualized in MRIcro.Regression maps revealed two clusters of brain activation for the speech versus silence contrast and five clusters for the noise versus silence contrast that were significantly correlated with the parental reports. These clusters included auditory and extra-auditory regions such as the middle temporal gyrus, supramarginal gyrus, precuneus, cingulate gyrus, middle frontal gyrus, subgyral, and middle occipital gyrus. Both positive and negative correlations were observed. Correlation values for the different clusters ranged from −0.90 to 0.95 and were significant at a corrected p value of The results of the present study suggest that (1) fMRI can be used to identify neuroimaging biomarkers of auditory and language performance before implantation and (2) activation in certain brain regions may be predictive of postimplant auditory and language performance as measured by parental observation/reports.

Breast cancer stem cells (BCSCs) contribute to intra-tumoral heterogeneity and therapeutic resistance. However, the binary concept of universal BCSCs co-existing with bulk tumor cells is over-simplified. Through single-cell RNA-sequencing, we found that Neu, PyMT, BRCA1-null, and 4T1 mammary tumors each corresponded to a spectrum of minimally overlapping cell differentiation states without a universal BCSC population. Instead, our analyses revealed that these tumors contained distinct lineage-specific BCSCs and this is reflective of the self-sustaining capabilities of lineage-specific stem/progenitor cells in the mammary epithelial hierarchy. By understanding the respective tumor hierarchies, we were able to identify CD14 as a BCSC marker in the Neu tumor. Additionally, single-cell breast cancer subtype stratification revealed the co-existence of multiple breast cancer subtypes within tumors. Collectively, our findings emphasize the need to account for lineage-specific BCSCs and the hierarchical composition within breast tumors, as these heterogenous sub-populations can have differential therapeutic susceptibilities.

OBJECTIVES Despite the positive effects of cochlear implantation, postimplant variability in speech perception and oral language outcomes is still difficult to predict. The aim of this study was to identify neuroimaging biomarkers of postimplant speech perception and oral language performance in children with hearing loss who receive a cochlear implant. The authors hypothesized positive correlations between blood oxygen level-dependent functional magnetic resonance imaging (fMRI) activation in brain regions related to auditory language processing and attention and scores on the Clinical Evaluation of Language Fundamentals-Preschool, Second Edition (CELF-P2) and the Early Speech Perception Test for Profoundly Hearing-Impaired Children (ESP), in children with congenital hearing loss. DESIGN Eleven children with congenital hearing loss were recruited for the present study based on referral for clinical MRI and other inclusion criteria. All participants were

Proposed a decomposed evolutionary algorithm for learning fuzzy cognitive map models.Applied the proposed method to infer causal networks from gene expression time series.Learned 300-node fuzzy cognitive maps while less than 40 nodes are common in the literature.Compared four stochastic optimization methods for learning fuzzy cognitive maps. Fuzzy cognitive maps have been widely used as abstract models for complex networks. Traditional ways to construct fuzzy cognitive maps rely on domain knowledge. In this paper, we propose to use fuzzy cognitive map learning algorithms to discover domain knowledge in the form of causal networks from data. More specifically, we propose to infer gene regulatory networks from gene expression data. Furthermore, a new efficient fuzzy cognitive map learning algorithm based on a decomposed genetic algorithm is developed to learn large scale networks. In the proposed algorithm, the simulation error is used as the objective function, while the model error is expected to be minimized. Experiments are performed to explore the feasibility of this approach. The high accuracy of the generated models and the approximate correlation between simulation errors and model errors suggest that it is possible to discover causal networks using fuzzy cognitive map learning. We also compared the proposed algorithm with ant colony optimization, differential evolution, and particle swarm optimization in a decomposed framework. Comparison results reveal the advantage of the decomposed genetic algorithm on datasets with small data volumes, large network scales, or the presence of noise.

Mucoid mucA22 Pseudomonas aeruginosa (PA) is an opportunistic lung pathogen of cystic fibrosis (CF) and chronic obstructive pulmonary disease (COPD) patients that is highly sensitive to acidified nitrite (A-NO2-). In this study, we first screened PA mutant strains for sensitivity or resistance to 20 mM A-NO2- under anaerobic conditions that represent the chronic stages of the aforementioned diseases. Mutants found to be sensitive to A-NO2- included PA0964 (pmpR, PQS biosynthesis), PA4455 (probable ABC transporter permease), katA (major catalase, KatA) and rhlR (quorum sensing regulator). In contrast, mutants lacking PA0450 (a putative phosphate transporter) and PA1505 (moaA2) were A-NO2- resistant. However, we were puzzled when we discovered that mucA22 mutant bacteria, a frequently isolated mucA allele in CF and to a lesser extent COPD, were more sensitive to A-NO2- than a truncated ΔmucA deletion (Δ157-194) mutant in planktonic and biofilm culture, as well as during a chronic murine lung infection. Subsequent transcriptional profiling of anaerobic, A-NO2--treated bacteria revealed restoration of near wild-type transcript levels of protective NO2- and nitric oxide (NO) reductase (nirS and norCB, respectively) in the ΔmucA mutant in contrast to extremely low levels in the A-NO2--sensitive mucA22 mutant. Proteins that were S-nitrosylated by NO derived from A-NO2- reduction in the sensitive mucA22 strain were those involved in anaerobic respiration (NirQ, NirS), pyruvate fermentation (UspK), global gene regulation (Vfr), the TCA cycle (succinate dehydrogenase, SdhB) and several double mutants were even more sensitive to A-NO2-. Bioinformatic-based data point to future studies designed to elucidate potential cellular binding partners for MucA and MucA22. Given that A-NO2- is a potentially viable treatment strategy to combat PA and other infections, this study offers novel developments as to how clinicians might better treat problematic PA infections in COPD and CF airway diseases.

Metabolomics is the comprehensive study of small molecule metabolites in biological systems. By assaying and analyzing thousands of metabolites in biological samples, it provides a whole picture of metabolic status and biochemical events happening within an organism and has become an increasingly powerful tool in the disease research. In metabolomics, it is common to deal with large amounts of data generated by nuclear magnetic resonance (NMR) and/or mass spectrometry (MS). Moreover, based on different goals and designs of studies, it may be necessary to use a variety of data analysis methods or a combination of them in order to obtain an accurate and comprehensive result. In this review, we intend to provide an overview of computational and statistical methods that are commonly applied to analyze metabolomics data. The review is divided into five sections. The first two sections will introduce the background and the databases and resources available for metabolomics research. The third section will briefly describe the principles of the two main experimental methods that produce metabolomics data: MS and NMR, followed by the fourth section that describes the preprocessing of the data from these two approaches. In the fifth and the most important section, we will review four main types of analysis that can be performed on metabolomics data with examples in metabolomics. These are unsupervised learning methods, supervised learning methods, pathway analysis methods and analysis of time course metabolomics data. We conclude by providing a table summarizing the principles and tools that we discussed in this review.

Aims Impaired sarcoplasmic reticulum (SR) Ca2+ cycling and depressed contractility, a hallmark of human and experimental heart failure, has been partially attributed to increased protein phosphatase 1 (PP-1) activity, associated with down-regulation of its endogenous inhibitor-1. The levels and activity of inhibitor-1 are reduced in failing hearts, contributing to dephosphorylation and inactivation of key calcium cycling proteins. Therefore, we investigated the mechanisms that mediate decreases in inhibitor-1 by post-transcriptional modification. Methods and Results Bioinformatics revealed that 17 human microRNAs may serve as modulators of inhibitor-1. However, real-time PCR analysis identified only one of these microRNAs, miR-765, as being increased in human failing hearts concomitant with decreased inhibitor-1 levels. Expression of miR-765 in HEK293 cells or mouse ventricular myocytes confirmed suppression of inhibitor-1 levels through binding of this miR-765 to the 3'-untranslated region of inhibitor-1 mRNA. To determine the functional significance of miR-765 in Ca2+ cycling, pri-miR-765 as well as a non-translated nucleotide sequence (miR-Ctrl) were expressed in adult mouse ventricular myocytes. The inhibitor-1 expression levels were decreased, accompanied by enhanced PP-1 activity in the miR-765 cardiomyocytes, and these reflected depressed contractile mechanics and Ca2+ transients, compared with the miR-Ctrl group. The depressive effects were associated with decreases in the phosphorylation of phospholamban and SR Ca2+ load. These miR-765 negative inotropic effects were abrogated in inhibitor-1-deficient cardiomyocytes, suggesting its apparent specificity for inhibitor-1. Conclusions miR-765 levels are increased in human failing hearts. Such increases may contribute to depressed cardiac function through reduced inhibitor-1 expression and enhanced PP-1 activity, associated with reduced SR Ca2+ load.

Motivation: Genes with indispensable functions are identified as essential; however, the traditional gene-level studies of essentiality have several limitations. In this study, we characterized gene essentiality from a new perspective of protein domains, the independent structural or functional units of a polypeptide chain. Results: To identify such essential domains, we have developed an Expectation–Maximization (EM) algorithm-based Essential Domain Prediction (EDP) Model. With simulated datasets, the model provided convergent results given different initial values and offered accurate predictions even with noise. We then applied the EDP model to six microbial species and predicted 1879 domains to be essential in at least one species, ranging 10–23% in each species. The predicted essential domains were more conserved than either non-essential domains or essential genes. Comparing essential domains in prokaryotes and eukaryotes revealed an evolutionary distance consistent with that inferred from ribosomal RNA. When utilizing these essential domains to reproduce the annotation of essential genes, we received accurate results that suggest protein domains are more basic units for the essentiality of genes. Furthermore, we presented several examples to illustrate how the combination of essential and non-essential domains can lead to genes with divergent essentiality. In summary, we have described the first systematic analysis on gene essentiality on the level of domains. Contact: huilu.bioinfo@gmail.com or Long.Lu@cchmc.org Supplementary Information: Supplementary data are available at Bioinformatics online.

Real-world data such as medical images and sensor measurements is usually high-dimensional and limited. Using such datasets directly in machine learning tasks can lead to poor generalization. Feature learning is a general approach for transforming high-dimensional data points to a representational space with lower dimensionality. Machine learning models can be trained efficiently with such representations. In this paper, a novel feature selection method based on multiple trained sparse auto-encoders (SAEs) is described. It works by selecting diverse, non-redundant features from multiple pinched SAEs with very narrow hidden layers, and then using these features in a more appropriately sized classifier without further feature tuning. The feature learning ability of the method is evaluated in a handwritten digits recognition task. Results show that this type of feature selection provides improved representations for a softmax classifier, and that using pinched SAEs produces results equal to or better than regular SAEs.

HDL has been shown to possess a variety of cardio-protective functions, including removal of excess cholesterol from the periphery, and inhibition of lipoprotein oxidation. It has been proposed that various HDL subparticles exist, each with distinct protein and lipid compositions, which may be responsible for HDL's many functions. We hypothesized that HDL functions will co-migrate with the operational lipoprotein subspecies when separated by gel filtration chromatography. Plasma from 10 healthy male donors was fractionated and the protein composition of the phospholipid containing fractions was analyzed by mass spectrometry (MS). Each fraction was evaluated for its proteomic content as well as its ability to promote cholesterol efflux and protect low density lipoprotein (LDL) from free radical oxidation. For each function, several peaks of activity were identified across the plasma size gradient. Neither cholesterol efflux or LDL antioxidation activity correlated strongly with any single protein across the fractions. However, we identified multiple proteins that had strong correlations (r values >0.7, p < 0.01) with individual peaks of activity. These proteins fell into diverse functional categories, including those traditionally associated with lipid metabolism, as well as alternative complement cascade, innate immunity and clotting cascades and immunoglobulins. Additionally, the phospholipid and cholesterol concentration of the fractions correlated strongly with cholesterol efflux (r = 0.95 and 0.82 respectively), whereas the total protein content of the fractions correlated best with antioxidant activity across all fractions (r = 0.746). Furthermore, two previously postulated subspecies (apoA-I, apoA-II and apoC-1; as well as apoA-I, apoC-I and apoJ) were found to have strong correlations with both cholesterol efflux and antioxidation activity. Up till now, very little has been known about how lipoprotein composition mediates functions like cholesterol efflux and antioxidation.

Networks, where nodes denote entities and links denote associations, provide a unified representation for a variety of complex systems, from social relationships to molecular interactions. In an era of big data, network analysis has been proved useful in biological applications such as predicting functions of proteins, guiding the design of wet-lab experiments, and discovering biomarkers of diseases. Driven by the availability of large-scale data sets and rapid development of bioinformatics’ tools, the research community has applied network analysis to define underlying causes of pediatric diseases. This will almost certainly lead to more effective strategies for prevention and treatment of diseases. In this chapter, we will introduce classic and the state-of-the-art network analysis methodologies, approaches and their applications. We then provide four examples of recent research, where network analysis is being applied in pediatrics. These include the identification of high-density lipoprotein particles that underlie the development of cardiovascular disease using protein-protein interaction networks, alternative splicing analysis by splicing interaction network, construction and network analysis of pediatric brain functional atlas, and disease relationship exploration using diagnosis association networks constructed by electronic health record.

Accurately predicting essential genes is important in many aspects of biology, medicine and bioengineering. In previous research, we have developed a machine learning based integrative algorithm to predict essential genes in bacterial species. This algorithm lends itself to two approaches for predicting essential genes: learning the traits from known essential genes in the target organism, or transferring essential gene annotations from a closely related model organism. However, for an understudied microbe, each approach has its potential limitations. The first is constricted by the often small number of known essential genes. The second is limited by the availability of model organisms and by evolutionary distance. In this study, we aim to determine the optimal strategy for predicting essential genes by examining four microbes with well-characterized essential genes. Our results suggest that, unless the known essential genes are few, learning from the known essential genes in the target organism usually outperforms transferring essential gene annotations from a related model organism. In fact, the required number of known essential genes is surprisingly small to make accurate predictions. In prokaryotes, when the number of known essential genes is greater than 2% of total genes, this approach already comes close to its optimal performance. In eukaryotes, achieving the same best performance requires over 4% of total genes, reflecting the increased complexity of eukaryotic organisms. Combining the two approaches resulted in an increased performance when the known essential genes are few. Our investigation thus provides key information on accurately predicting essential genes and will greatly facilitate annotations of microbial genomes.

The low prevalence rate of orphan diseases (OD) requires special combined efforts to improve diagnosis, prevention, and discovery of novel therapeutic strategies. To identify and investigate relationships based on shared genes or shared functional features, we have conducted a bioinformatic-based global analysis of all orphan diseases with known disease-causing mutant genes. Starting with a bipartite network of known OD and OD-causing mutant genes and using the human protein interactome, we first construct and topologically analyze three networks: the orphan disease network, the orphan disease-causing mutant gene network, and the orphan disease-causing mutant gene interactome. Our results demonstrate that in contrast to the common disease-causing mutant genes that are predominantly nonessential, a majority of orphan disease-causing mutant genes are essential. In confirmation of this finding, we found that OD-causing mutant genes are topologically important in the protein interactome and are ubiquitously expressed. Additionally, functional enrichment analysis of those genes in which mutations cause ODs shows that a majority result in premature death or are lethal in the orthologous mouse gene knockout models. To address the limitations of traditional gene-based disease networks, we also construct and analyze OD networks on the basis of shared enriched features (biological processes, cellular components, pathways, phenotypes, and literature citations). Analyzing these functionally-linked OD networks, we identified several additional OD-OD relations that are both phenotypically similar and phenotypically diverse. Surprisingly, we observed that the wiring of the gene-based and other feature-based OD networks are largely different; this suggests that the relationship between ODs cannot be fully captured by the gene-based network alone.

Rapid and accurate identification of new essential genes in under-studied microorganisms will significantly improve our understanding of how a cell works and the ability to re-engineer microorganisms. However, predicting essential genes across distantly related organisms remains a challenge. Here, we present a machine learning-based integrative approach that reliably transfers essential gene annotations between distantly related bacteria. We focused on four bacterial species that have well-characterized essential genes, and tested the transferability between three pairs among them. For each pair, we trained our classifier to learn traits associated with essential genes in one organism, and applied it to make predictions in the other. The predictions were then evaluated by examining the agreements with the known essential genes in the target organism. Ten-fold cross-validation in the same organism yielded AUC scores between 0.86 and 0.93. Cross-organism predictions yielded AUC scores between 0.69 and 0.89. The transferability is likely affected by growth conditions, quality of the training data set and the evolutionary distance. We are thus the first to report that gene essentiality can be reliably predicted using features trained and tested in a distantly related organism. Our approach proves more robust and portable than existing approaches, significantly extending our ability to predict essential genes beyond orthologs.

SummaryA remarkable feature of development is its reproducibility, the ability to correct embryo-to-embryo variations and instruct precise patterning. In Drosophila, embryonic patterning along the anterior-posterior axis is controlled by the morphogen gradient Bicoid (Bcd). In this article, we describe quantitative studies of the native Bcd gradient and its target Hunchback (Hb). We show that the native Bcd gradient is highly reproducible and is itself scaled with embryo length. While a precise Bcd gradient is necessary for precise Hb expression, it still has positional errors greater than Hb expression. We describe analyses further probing mechanisms for Bcd gradient scaling and correction of its residual positional errors. Our results suggest a simple model of a robust Bcd gradient sufficient to achieve scaled and precise activation of its targets. The robustness of this gradient is conferred by its intrinsic properties of “self-correcting” the inevitable input variations to achieve a precise and reproducible output.

Introduction We developed a machine learning model to predict whether or not a cochlear implant (CI) candidate will develop effective language skills within 2 years after the CI surgery by using the pre-implant brain fMRI data from the candidate. Methods The language performance was measured 2 years after the CI surgery by the Clinical Evaluation of Language Fundamentals-Preschool, Second Edition (CELF-P2). Based on the CELF-P2 scores, the CI recipients were designated as either effective or ineffective CI users. For feature extraction from the fMRI data, we constructed contrast maps using the general linear model, and then utilized the Bag-of-Words (BoW) approach that we previously published to convert the contrast maps into feature vectors. We trained both supervised models and semi-supervised models to classify CI users as effective or ineffective. Results Compared with the conventional feature extraction approach, which used each single voxel as a feature, our BoW approach gave rise to much better performance for the classification of effective versus ineffective CI users. The semi-supervised model with the feature set extracted by the BoW approach from the contrast of speech versus silence achieved a leave-one-out cross-validation AUC as high as 0.97. Recursive feature elimination unexpectedly revealed that two features were sufficient to provide highly accurate classification of effective versus ineffective CI users based on our current dataset. Conclusion We have validated the hypothesis that pre-implant cortical activation patterns revealed by fMRI during infancy correlate with language performance 2 years after cochlear implantation. The two brain regions highlighted by our classifier are potential biomarkers for the prediction of CI outcomes. Our study also demonstrated the superiority of the semi-supervised model over the supervised model. It is always worthwhile to try a semi-supervised model when unlabeled data are available.

Genes with indispensable functions are identified as essential; however, the traditional gene-level perspective of essentiality has several limitations. We hypothesized that protein domains, the independent structural or functional units of a polypeptide chain, are responsible for gene essentiality. If the essentiality of domains is known, the essential genes could be identified. To find such essential domains, we have developed an EM algorithm-based Essential Domain Prediction (EDP) Model. With simulated datasets, the model provided convergent results given different initial values and offered accurate predictions even with noise. We then applied the EDP model to six microbes and predicted 3,450 domains to be essential in at least one species, ranging 8-24 % in each species.

Genomic data integration—the process of statistically combining diverse sources of information from functional genomics experiments to make large-scale predictions—is becoming increasingly prevalent. One might expect that this process should become progressively more powerful with the integration of more evidence. Here, we explore the limits of genomic data integration, assessing the degree to which predictive power increases with the addition of more features. We focus on a predictive context that has been extensively investigated and benchmarked in the past—the prediction of protein–protein interactions in yeast. We start by using a simple Naive Bayes classifier for integrating diverse sources of genomic evidence, ranging from coexpression relationships to similar phylogenetic profiles. We expand the number of features considered for prediction to 16, significantly more than previous studies. Overall, we observe a small, but measurable improvement in prediction performance over previous benchmarks, based on four strong features. This allows us to identify new yeast interactions with high confidence. It also allows us to quantitatively assess the inter-relations amongst different genomic features. It is known that subtle correlations and dependencies between features can confound the strength of interaction predictions. We investigate this issue in detail through calculating mutual information. To our surprise, we find no appreciable statistical dependence between the many possible pairs of features. We further explore feature dependencies by comparing the performance of our simple Naive Bayes classifier with a boosted version of the same classifier, which is fairly resistant to feature dependence. We find that boosting does not improve performance, indicating that, at least for prediction purposes, our genomic features are essentially independent. In summary, by integrating a few (i.e., four) good features, we approach the maximal predictive power of current genomic data integration; moreover, this limitation does not reflect (potentially removable) inter-relationships between the features.

BackgroundAspergillus fumigatus (Af) is a ubiquitous and opportunistic pathogen capable of causing acute, invasive pulmonary disease in susceptible hosts. Despite current therapeutic options, mortality associated with invasive Af infections remains unacceptably high, increasing 357% since 1980. Therefore, there is an urgent need for the development of novel therapeutic strategies, including more efficacious drugs acting on new targets. Thus, as noted in a recent review, "the identification of essential genes in fungi represents a crucial step in the development of new antifungal drugs". Expanding the target space by rapidly identifying new essential genes has thus been described as "the most important task of genomics-based target validation". ResultsIn previous research, we were the first to show that essential gene annotation can be reliably transferred between distantly related four Prokaryotic species. In this study, we extend our machine learning approach to the much more complex Eukaryotic fungal species. A compendium of essential genes is predicted in Af by transferring known essential gene annotations from another filamentous fungus Neurospora crassa. This approach predicts essential genes by integrating diverse types of intrinsic and context-dependent genomic features encoded in microbial genomes. The predicted essential datasets contained 1674 genes. We validated our results by comparing our predictions with known essential genes in Af, comparing our predictions with those predicted by homology mapping, and conducting conditional expressed alleles. We applied several layers of filters and selected a set of potential drug targets from the predicted essential genes. Finally, we have conducted wet lab knockout experiments to verify our predictions, which further validates the accuracy and wide applicability of the machine learning approach. ConclusionsThe approach presented here significantly extended our ability to predict essential genes beyond orthologs and made it possible to predict an inventory of essential genes in Eukaryotic fungal species, amongst which a preferred subset of suitable drug targets may be selected. By selecting the best new targets, we believe that resultant drugs would exhibit an unparalleled clinical impact against a naive pathogen population. Additional benefits that a compendium of essential genes can provide are important information on cell function and evolutionary biology. Furthermore, mapping essential genes to pathways may also reveal critical check points in the pathogen's metabolism. Finally, this approach is highly reproducible and portable, and can be easily applied to predict essential genes in many more pathogenic microbes, especially those unculturable.

Detecting brain structural changes from magnetic resonance (MR) images can facilitate early diagnosis and treatment of neurological and psychiatric diseases. Many existing methods require an accurate deformation registration, which is difficult to achieve and therefore prevents them from obtaining high accuracy. We develop a novel local feature based support vector machine (SVM) approach to detect brain structural changes as potential biomarkers. This approach does not require deformation registration and thus is less influenced by artifacts such as image distortion. We represent the anatomical structures based on scale invariant feature transform (SIFT). Likelihood scores calculated using feature-based morphometry is used as the criterion to categorize image features into three classes (healthy, patient and noise). Regional SVMs are trained to classify the three types of image features in different brain regions. Only healthy and patient features are used to predict the disease status of new brain images. An ensemble classifier is built from the regional SVMs to obtain better prediction accuracy. We apply this approach to 3D MR images of Alzheimer's disease, Parkinson's disease and bipolar disorder. The classification accuracy ranges between 70% and 87%. The highly predictive disease-related regions, which represent significant anatomical differences between the healthy and diseased, are shown in heat maps. The common and disease-specific brain regions are identified by comparing the highly predictive regions in each disease. All of the top-ranked regions are supported by literature. Thus, this approach will be a promising tool for assisting automatic diagnosis and advancing mechanism studies of neurological and psychiatric diseases.

Gene Regulatory Networks (GRNs) represent the causal relations among the genes and provide insight on the cellular functions and the mechanism of the diseases. GRNs can be inferred from gene expression data by a number of algorithms, e.g. Boolean networks, Bayesian networks, and differential equations. While reliable inference of GRNs is still an open problem, new algorithms need to be developed. Fuzzy Cognitive Maps (FCMs) is used to represent GRNs in this paper. Most of the FCM learning algorithms are able to learn FCMs with less than 40 nodes. A new algorithm that is able to learn FCMs with more than 100 nodes is proposed. The proposed method is based on Ant Colony Optimization (ACO). A decomposed approach is proposed to reduce the dimension of the problem; therefore the FCM learning algorithm is more scalable (the dimension of the problem to be solved in one ACO run equals to the number of nodes or genes). The proposed approach is tested on data from DREAM project. The experiment results suggest the proposed approach outperforms several other algorithms.

Fuzzy Cognitive Maps (FCMs) can represent causal relations among concepts in the form of directed weighted graphs. Learning FCMs from data is a difficult task because of the large number of candidate FCMs. The goal of learning FCMs is to find good weights for the FCM so that the response of the learned FCM should be as close to the observed data as possible. The number of variables to be optimized grows quadratically to the number of concepts. Furthermore, the non-linear numerical optimization problem is also a NP-hard problem. A FCM development algorithm based on Ant Colony Optimization is proposed in this paper. A local search procedure is designed to improve the performance of the learning algorithm. The FCM learning problem is formulated as searching for weights of the FCM so that the FCM has a response sequence as close to the observed data as possible. Simulated data generated from several FCMs with different number of concepts is used to test the ACO-based learning algorithm. The proposed algorithm outperforms other data-driven FCM learning algorithms such as non-linear Hebbian learning, data-driven non-linear Hebbian learning and real-coded genetic algorithm in several experiment scenarios.

Networks, where nodes/vertices denote entities and links/edges denote associations, provide a unified representation for a variety of complex systems, from social relationships to molecular interactions. In general, network analysis has proved useful in applications such as prediction of protein function, guiding the design of wet-lab experiments, and discovery of biomarkers of disease. Driven by the availability of large-scale data sets and appropriate informatics’ tools, the research community has begun to apply network analysis of these data to define underlying causes of pediatric diseases. This will almost certainly lead to more effective strategies for prevention and treatment. In this chapter, we will introduce classic and the state-of-the-art network analysis methodologies, approaches and their applications. We then provide three examples of recent research, where network analysis is being applied in pediatrics. These include identification of targets for development of new drugs to treat Pseudomonas infections, protein interactions in high-density lipoprotein particles that underlie development of cardiovascular disease, and analysis of networks in brain that may underlie conditions such as Attention Deficit Hyperactivity Disorder.

Pseudomonas aeruginosa (PA) is a ubiquitous opportunistic pathogen that is capable of causing highly problematic, chronic infections in cystic fibrosis and chronic obstructive pulmonary disease patients. With the increased prevalence of multi-drug resistant PA, the conventional "one gene, one drug, one disease" paradigm is losing effectiveness. Network pharmacology, on the other hand, may hold the promise of discovering new drug targets to treat a variety of PA infections. However, given the urgent need for novel drug target discovery, a PA protein-protein interaction (PPI) network of high accuracy and coverage, has not yet been constructed. In this study, we predicted a genome-scale PPI network of PA by integrating various genomic features of PA proteins/genes by a machine learning-based approach. A total of 54,107 interactions covering 4,181 proteins in PA were predicted. A high-confidence network combining predicted high-confidence interactions, a reference set and verified interactions that consist of 3,343 proteins and 19,416 potential interactions was further assembled and analyzed. The predicted interactome network from this study is the first large-scale PPI network in PA with significant coverage and high accuracy. Subsequent analysis, including validations based on existing small-scale PPI data and the network structure comparison with other model organisms, shows the validity of the predicted PPI network. Potential drug targets were identified and prioritized based on their essentiality and topological importance in the high-confidence network. Host-pathogen protein interactions between human and PA were further extracted and analyzed. In addition, case studies were performed on protein interactions regarding anti-sigma factor MucA, negative periplasmic alginate regulator MucB, and the transcriptional regulator RhlR. A web server to access the predicted PPI dataset is available at http://research.cchmc.org/PPIdatabase/.

Expanded polyglutamine (polyQ) tract in the human TATA-box-binding protein (hTBP) causes the neurodegenerative disease spinocerebellar ataxia 17 (SCA17). To investigate the pathological effects of polyQ expansion, we established a SCA17 model in Drosophila. Similar to SCA17 patients, transgenic flies expressing a mutant hTBP protein with an expanded polyQ tract (hTBP80Q) exhibit progressive neurodegeneration, late-onset locomotor impairment and shortened lifespan. Microarray analysis reveals that hTBP80Q causes widespread and time-dependent transcriptional dysregulation in Drosophila. In a candidate screen for genetic modifiers, we identified RBP-J/Su(H), a transcription factor that contains Q/N-rich domains and participates in Notch signaling. Knockdown of Su(H) by RNAi further enhances hTBP80Q-induced eye defects, whereas overexpression of Su(H) suppresses such defects. While the Su(H) transcript level is not significantly altered in hTBP80Q-expressing flies, genes that contain Su(H)-binding sites are among those that are dysregulated. We further show that hTBP80Q interacts more efficiently with Su(H) than wild-type hTBP, suggesting that a reduction in the fraction of Su(H) available for its normal cellular functions contributes to hTBP80Q-induced phenotypes. While the Notch signaling pathway has been implicated in several neurological disorders, our study suggests a possibility that the activity of its nuclear component RBP-J/Su(H) may modulate the pathological progression in SCA17 patients.

Transcriptional regulatory network identification is both a fundamental challenge in systems biology and an important practical application of data mining and machine learning. In this study, we propose a semi-supervised learning-based integrative scoring approach to tackle this challenge and predict transcriptional regulations. Our approach out-performs a state-of-the-art label propagation method and reaches AUC scores above 0.96 for three datasets from microarray experiments in the validation. A map of the transcriptional regulatory network controlling lung surfactant homeostasis was constructed. The predicted and prioritized transcriptional regulations were further validated through experimental verifications. Many other predicted novel regulations may serve as candidates for future experimental investigations.

Background A molecular network perspective forms the foundation of systems biology. A common practice in analyzing protein-protein interaction (PPI) networks is to perform network analysis on a conglomerate network that is an assembly of all available binary interactions in a given organism from diverse data sources. Recent studies on network dynamics suggested that this approach might have ignored the dynamic nature of context-dependent molecular systems. Results In this study, we employed a network stratification strategy to investigate the validity of the current network analysis on conglomerate PPI networks. Using the genome-scale tissue- and condition-specific proteomics data in Arabidopsis thaliana, we present here the first systematic investigation into this question. We stratified a conglomerate A. thaliana PPI network into three levels of context-dependent subnetworks. We then focused on three types of most commonly conducted network analyses, i.e., topological, functional and modular analyses, and compared the results from these network analyses on the conglomerate network and five stratified context-dependent subnetworks corresponding to specific tissues. Conclusions We found that the results based on the conglomerate PPI network are often significantly different from those of context-dependent subnetworks corresponding to specific tissues or conditions. This conclusion depends neither on relatively arbitrary cutoffs (such as those defining network hubs or bottlenecks), nor on specific network clustering algorithms for module extraction, nor on the possible high false positive rates of binary interactions in PPI networks. We also found that our conclusions are likely to be valid in human PPI networks. Furthermore, network stratification may help resolve many controversies in current research of systems biology.

Background Patterning along the anterior-posterior (A-P) axis in Drosophila embryos is instructed by the morphogen gradient of Bicoid (Bcd). Despite extensive studies of this morphogen, how embryo geometry may affect gradient formation and target responses has not been investigated experimentally. Results In this report, we systematically compare the Bcd gradient profiles and its target expression patterns on the dorsal and ventral sides of the embryo. Our results support a hypothesis that proper distance measurement and the encoded positional information of the Bcd gradient are along the perimeter of the embryo. Our results also reveal that the dorsal and ventral sides of the embryo have a fundamentally similar relationship between Bcd and its target Hunchback (Hb), suggesting that Hb expression properties on the two sides of the embryo can be directly traced to Bcd gradient properties. Our 3-D simulation studies show that a curvature difference between the two sides of an embryo is sufficient to generate Bcd gradient properties that are consistent with experimental observations. Conclusions The findings described in this report provide a first quantitative, experimental evaluation of embryo geometry on Bcd gradient formation and target responses. They demonstrate that the physical features of an embryo, such as its shape, are integral to how pattern is formed.

In the post-genomic era of systems biology, the cell itself can be viewed as a complex network of interacting proteins, nucleic acids, and other biomolecules (Hartwell et al., 1999; Eisenberg et al., 2000). Graph representation adopted from mathematics and computer science has been widely applied to describe various molecular systems including protein interaction maps, metabolites and reactions, transcriptional regulation maps, signal transduction pathways, and functional association networks (Barabasi and Oltvai, 2004; Girvan and Newman, 2002; Tong et al., 2004; Balazsi et al., 2005). Applications based on network analysis have been proven useful in areas such as predicting protein functions, identifying targets for structural genomics, facilitating drug discovery and design, and expe­ diting novel biomarker identification (Sharan et al., 2007; Chuang et al., 2007; Hopkins, 2008; Huang et al., 2008). In the first four sections of this chapter, we will review four types of commonly conducted network analyses on large-scale molecular networks and their applications: (1) Topology analysis focuses on the characterization of global network structures using quantitative measures, or network statistics. The applications of topology analysis include important node identification

Background Bicoid (Bcd) is a Drosophila morphogenetic protein responsible for patterning the anterior structures in embryos. Recent experimental studies have revealed important insights into the behavior of this morphogen gradient, making it necessary to develop a model that can recapitulate the biological features of the system, including its dynamic and scaling properties. Methodology/Principal Findings We present a biologically realistic 2-D model of the dynamics of the Bcd gradient in Drosophila embryos. This model is based on equilibrium binding of Bcd molecules to non-specific, low affinity DNA sites throughout the Drosophila genome. It considers both the diffusion media within which the Bcd gradient is formed and the dynamic and other relevant properties of bcd mRNA from which Bcd protein is produced. Our model recapitulates key features of the Bcd protein gradient observed experimentally, including its scaling properties and the stability of its nuclear concentrations during development. Our simulation model also allows us to evaluate the effects of other biological activities on Bcd gradient formation, including the dynamic redistribution of bcd mRNA in early embryos. Our simulation results suggest that, in our model, Bcd protein diffusion is important for the formation of an exponential gradient in embryos. Conclusions/Significance The 2-D model described in this report is a simple and versatile simulation procedure, providing a quantitative evaluation of the Bcd gradient system. Our results suggest an important role of Bcd binding to non-specific, low-affinity DNA sites in proper formation of the Bcd gradient in our model. They demonstrate that highly complex biological systems can be effectively modeled with relatively few parameters.

Structural genomics provides an important approach for characterizing and understanding systems biology. As a step toward better integrating protein three-dimensional (3D) structural information in cancer systems biology, we have constructed a Human Cancer Pathway Protein Interaction Network (HCPIN) by analysis of several classical cancer-associated signaling pathways and their physical protein-protein interactions. Many well known cancer-associated proteins play central roles as "hubs" or "bottlenecks" in the HCPIN. At least half of HCPIN proteins are either directly associated with or interact with multiple signaling pathways. Although some 45% of residues in these proteins are in sequence segments that meet criteria sufficient for approximate homology modeling (Basic Local Alignment Search Tool (BLAST) E-value10(-6)), only approximately 20% of residues in these proteins are structurally covered using high accuracy homology modeling criteria (i.e. BLAST E-value10(-6) and at least 80% sequence identity) or by actual experimental structures. The HCPIN Website provides a comprehensive description of this biomedically important multipathway network together with experimental and homology models of HCPIN proteins useful for cancer biology research. To complement and enrich cancer systems biology, the Northeast Structural Genomics Consortium is targeting1000 human proteins and protein domains from the HCPIN for sample production and 3D structure determination. The long range goal of this effort is to provide a comprehensive 3D structure-function database for human cancer-associated proteins and protein complexes in the context of their interaction networks. The network-based target selection (BioNet) approach described here is an example of a general strategy for targeting co-functioning proteins by structural genomics projects.

A global protein survey is needed to gain systems-level insights into mammalian cell signaling and information flow. Human Jurkat T leukemic cells are one of the most important model systems for T cell signaling study, but no comprehensive proteomics survey has been carried out in this cell type. In the present study we combined subcellular fractionation, multiple protein enrichment methods, and replicate tandem mass spectrometry analyses to determine the protein expression pattern in a single Jurkat cell type. The proteome dataset was evaluated by comparison with the genome-wide mRNA expression pattern in the same cell type. A total of 5381 proteins were identified by mass spectrometry with high confidence. Rigorous comparison of RNA and protein expression afforded removal of the false positive identifications and redundant entries but rescued the proteins identified by a single high scoring peptide, resulting in the final identification of 6471 unique gene products among which 98% of the corresponding transcripts were detected with high probability. Using hierarchical clustering of the protein expression patterns in five subcellular fractions (cytosol, light membrane, heavy membrane, mitochondria, and nuclei), the primary subcellular localization of 2241 proteins was assigned with high confidence including 792 previously uncharacterized proteins. This proteome landscape can serve as a useful platform for systems-level understanding of organelle composition and cellular functions in human T cells.