Combining multiple positive training sets to generate confidence scores for protein-protein interactions

Motivation: High-throughput experimental and computational methodsare generating a wealth of protein–protein interactiondata for a variety of organisms. However, data produced by currentstate-of-the-art methods include many false positives, whichcan hinder the analyses needed to derive biological insights.One way to address this problem is to assign confidence scoresthat reflect the reliability and biological significance ofeach interaction. Most previously described scoring methodsuse a set of likely true positives to train a model to scoreall interactions in a dataset. A single positive training set,however, may be biased and not representative of true interactionspace. Results: We demonstrate a method to score protein interactionsby utilizing multiple independent sets of training positivesto reduce the potential bias inherent in using a single trainingset. We used a set of benchmark yeast protein interactions toshow that our approach outperforms other scoring methods. Ourapproach can also score interactions across data types, whichmakes it more widely applicable than many previously proposedmethods. We applied the method to protein interaction data fromboth Drosophila melanogaster and Homo sapiens. Independent evaluationsshow that the resulting confidence scores accurately reflectthe biological significance of the interactions. Contact: rfinley{at}wayne.edu Supplementary information: Supplementary data are availableat Bioinformatics Online. Associate Editor: Burkhard Rost     

Estimating equations for spatially correlated data in multi-dimensional space

We use the quasilikelihood concept to propose an estimatingequation for spatial data with correlation across the studyregion in a multi-dimensional space. With appropriate mixingconditions, we develop a central limit theorem for a randomfield under various L_p metrics. The consistency and asymptoticnormality of quasilikelihood estimators can then be derived.We also conduct simulations to evaluate the performance of theproposed estimating equation, and a dataset from East LansingWoods is used to illustrate the method.     

Semiparametric regression estimation in the presence of dependent censoring

We propose a semiparametric estimation procedure for estimatingthe regression of an outcome Y, measured at the end of a fixedfollow-up period, on baseline explanatory variables X, measuredprior to start of follow-up, in the presence of dependent censoringgiven X. The proposed estimators are consistent when the dataare ‘missing at random’ but not ‘missing completelyat random’ (Rubin, 1976), and do not require full specificationof the complete data likelihood. Specifically, we assume thatthe probability of censoring at time t is independent of theoutcome Y conditional on the recorded history up to t of a vectorof time-dependent covariates that are correlated with Y. Ourestimators can be used to adjust for dependent censoring andnonrandom noncompliance in randomised trials studying the effectof a treatment on the mean of a response variable of interest.Even with independent censoring, our methods allow the investigatorto increase efficiency by exploiting the correlation of theoutcome with a vector of time-dependent covariates.     

持续治理飞蝗灾害的新对策

该文通过对飞蝗Locusta migratoria为害特性的描述以及本世纪我国飞蝗防治历史的回顾,对各个时期的飞蝗治理策略的特点及不足进行了分析,并在总结我国近几年飞蝗生物治理理论研究与实践的基础上,提出了21世纪我国飞蝗生物治理的对策,即逐步扩大采用生物防治措施,增强如蝗虫微孢子虫生物防治制剂及天敌的控制作用,在飞蝗虫口密度中等或较低时,采用生物防治制剂为主,阻止或延缓中、低密度的飞蝗向高密度群居型的发展进程;在飞蝗虫口密度高时,可采用化防(昆虫生长调节剂等)与生防(如微孢子虫)配合使用,以迅速压低虫口密度,防治其迁飞为害,同时也可使蝗虫微孢子虫疾病长期流行于蝗群中,抑制飞蝗种群数量的增长。并积极探讨信息化合物对飞蝗行为的调控作用。     

A New Model Relating Crop Yield and Plant Arrangement

A new model is proposed which relates the weight of plants totheir spatial arrangement. The weight of each plant is calculatedas the integral of the function f(r) = L(cr² + 1)^–2 overan area allocated to it, r being distance from the plant, withL and c parameters to be specified. The model is thus concise,general, in that it can be used to describe the effects of anyspatial arrangement on plant weight, and the parameters L andc have a biological interpretation. It is also consistent withthe commonly-used relationship between plant weight (w) anddensity (p), w^–1 = a+bp. We show for carrots (Daucus carota L.) and red beet (Beta vulgarisL.), that the mean weights fitted by the model agree as wellwith the experimentally observed mean plot weight as those fittedby more complex models with more parameters, some of which arenot as general. We show also that the parameter c can be predictedfrom the time from sowing to harvest, with good results whentested on sets of data independent of those to which the modelhad been fitted. The assumptions on which the model is based,its application, and extensions to it are discussed. Crop yield, plant density, plant arrangement, carrot, Daucus carota L., red beet, Beta vulgaris L., soya bean, Glycine max L., mathematical model     

Zooplankton in the Ligurian Sea: Part II. Exploration of their physical and biological forcing functions during summer 2000

A survey of the biological and physical oceanography of theLigurian Sea was conducted in the late summer of 2000. Forty-onestations were sampled for nutrients, oxygen, fluorescence andhydrographic information. Acoustic backscatter measurementswere used to estimate abundance of small (<5 mm) zooplanktonbiovolume versus depth and the distribution of northern krill,Meganyctiphanes norvegica. Net-tow and underwater video datawere collected to identify the zooplankton present. These datawere used to analyze the Ligurian Sea ecosystem for physicaland biological linkages that control zooplankton abundance anddistribution. Results are compared with those from a similarstudy conducted in 1999. Hydrographic sampling showed a domeof dense water in the southwestern middle of the basin. Thehighest chlorophyll a (Chl a) concentrations were measured inthis area, while small zooplankton biovolume was evenly distributedthroughout the survey. Integrated values of Chl a and smallzooplankton biovolume in 2000 were greater than in 1999. Meganyctiphanesnorvegica, siphonophores and salps were the dominant componentsof the macrozooplankton population in the upper 200 m. In thesampled depth strata, siphonophore abundance did not changeduring the day, while M. norvegica were only caught at night.Acoustic backscatter data show that higher densities of M. norvegicaoccurred in deeper water and in the western and southwesternareas of the Ligurian Sea.     

The Photosynthesis and Cell Shape Rhythms can be Naturally Uncoupled from the Biological Clock in Euglena gracilis

A naturally occurring population of Euglena (Klebs Strain Z)cells, with unusual biological clock properties, has been isolated.The photosynthesis reactions, which are usually controlled bythe biological clock, are uncoupled from the clock in the newpopulation. The rate of oxygen evolution is influenced predominantlyby the environmental growth parameters instead of the biologicalclock. In addition, the rhythm in cellular shape has a differenttiming from the control population and can be temporarily uncoupledfrom the clock by lowering the light intensity used for growth Key words: Biological clock, photosynthesis, Euglena     

Interactions Between Physical and Biological Constraints in the Structure of the Inflorescences of the Araceae

A study of the inflorescences ofMonsteraandAnthuriumwas usedto establish a relationship between biological and physicalconstraints for the structure of plant organs. The physicalconstraint between flowers in the compact inflorescences ofAnthuriumandMonsteraisexpressed by Aboav-Weaire's law. The application of this lawto inflorescences indicates a linear relationship between thenumber of sides of a flower and the number of sides of neighbouringflowers. However, the slope of this straight line is significantlyhigher forAnthuriumandMonsterathan that expected in theory.This deviation from the law is attributable to a biologicalcause that can be estimated using Aboav-Weaire's law. Actingalone, the biological constraint tends to produce four-sidedflowers. The equilibrium between biological and physical constraintsreduces the number of sides per flower from six (theoreticalvalue) to 5.9 (inAnthurium) or 5.8 (inMonstera) with a varianceof the measures less than that expected in theory. Furthermore,when flower density in an inflorescence increases (towards themiddle of the inflorescence inMonsteraand towards the lowersection forAnthurium) the number of sides approaches six (i.e.the physical constraint dominates). When flower density decreases(towards the top of the inflorescence) the number of sides approaches5.5 (i.e. the biological constraint dominates). The geometryof the inflorescences ofAnthuriumandMonsterais the result ofthe joint action of biological and physical constraints.Copyright1998 Annals of Botany Company Monstera,Anthurium, Araceae, Aboav-Weaire, inflorescence, constraint, flower.     

Analysis of Organ Specificity of a Low Temperature Responsive Gene Family in Rye (Secale cereale L.)

The barley (Hordeum vulgare L.) low temperature responsive geneblt14 was used as a probe, to isolate two different cognateclones (rlt1412; rlt1421) from a rye (Secale cereale L.) cDNAlibrary prepared from low temperature-treated (6°C day/2°C night) shoot meristems of the cultivar, Puma. Northernblot analysis revealed that low temperature expression of rlt1412is highest in root tissues whereas, rlt1421 shows greatest mRNAaccumulation in mature leaf tissues. There is a relationshipbetween the steady-state levels of these mRNA species and thefrost hardiness of Puma (North American cultivar) and Rhayader(UK cultivar) such that the expression ofboth genes is higherin the more frost hardy cultivar, Puma, compared with Rhayader. DNA and predicted amino acid sequence analysis indicated thatthe rye and barley clones encode small proteins with consensusN-terminal signal sequences whose biological function is atpresent unknown. The relevant sequences are lodged in the EMBL data base. Key words: Rye, cold, cDNA, organ specificity, low temperature genes     

Selection and control of Deborah numbers in plankton ecology

The Deborah number (De) is widely used to characterize processestaking place in deforming continua. De=(the time scale of aprocess)/(the time scale of deformation). When De >>lthe process thus takes place in a functionally fluid medium,but when De <<1 the regime is functionally solid. De hasbeen used to refine concepts in three pelagic processes. Dispersionof dividing cells may be characterized by De, and may be regulatedby means of secretions. Dispersion of microzones by diffusionand shear is characterized. The characteristic time of microzonesis shown to depend on the concentration. Because microzonessmear Out along the shear, to prevent nutrient-seekers and predatorsusing them as scent trails, organisms may convolute their microzonesby swimming, particularly across the shear. In a predator-preymodel, it has been shown that when De, (shear rate) (time takento swim radius of detection sphere), >2.6, not all the perceivedprey is accessible. More economical hunting strategies and thoseallowing access to more of the perceived prey, require bettersensory and navigational abilities. When De >2.6, the predatorwill perceive a greater flux of accessible prey when it swimsacross the shear than when it swims in the other two dimensions.De may help to understand many more biological processes indeforming media.     

MMG: a probabilistic tool to identify submodules of metabolic pathways

Motivation: A fundamental task in systems biology is the identificationof groups of genes that are involved in the cellular responseto particular signals. At its simplest level, this often reducesto identifying biological quantities (mRNA abundance, enzymeconcentrations, etc.) which are differentially expressed intwo different conditions. Popular approaches involve using t-teststatistics, based on modelling the data as arising from a mixturedistribution. A common assumption of these approaches is thatthe data are independent and identically distributed; however,biological quantities are usually related through a complex(weighted) network of interactions, and often the more pertinentquestion is which subnetworks are differentially expressed,rather than which genes. Furthermore, in many interesting cases(such as high-throughput proteomics and metabolomics), onlyvery partial observations are available, resulting in the needfor efficient imputation techniques. Results: We introduce Mixture Model on Graphs (MMG), a novelprobabilistic model to identify differentially expressed submodulesof biological networks and pathways. The method can easily incorporateinformation about weights in the network, is robust againstmissing data and can be easily generalized to directed networks.We propose an efficient sampling strategy to infer posteriorprobabilities of differential expression, as well as posteriorprobabilities over the model parameters. We assess our methodon artificial data demonstrating significant improvements overstandard mixture model clustering. Analysis of our model resultson quantitative high-throughput proteomic data leads to theidentification of biologically significant subnetworks, as wellas the prediction of the expression level of a number of enzymes,some of which are then verified experimentally. Availability: MATLAB code is available from http://www.dcs.shef.ac.uk/~guido/software.html Contact: guido{at}dcs.shef.ac.uk Supplementary information: Supplementary data are availableat Bioinformatics online. Associate Editor: Jonathan Wren     

A Comparative Study of the Biological Activities of Two Molecular Species of Chloroplast-type Ferredoxins

Pairs of two molecular species of soluble chloroplast-type ferredoxins(Fd I and Fd II) from Nostoc muscorum and Aphanothece sacrumwere used to examine and compare the abilities of ferredoxinto substitute for spinach ferredoxin in the photoreduction ofNADP⁺ by spinach chloroplasts or N. muscorum membrane fragmentsand to link the reducing power of illuminated spinach chloroplaststo the Bacillus polymyxa nitrogenase system. Ferredoxins II of Nostoc and Aphanothece showed rather low activitiesin NADP⁺ photoreduction and nitrogenase system with spinachchloroplasts as the photosensitizer, compared to other ferredoxins.However, there was no difference between two ferredoxins (FdI and Fd II) from Nostoc in NADP⁺ photoreduction by photosyntheticmembrane fragments prepared from the same organism, N. muscorum. The biological significance of two molecular species of ferredoxinsin one organism could be ascribed to the different contributionof each ferredoxin to certain biological reactions in whichferredoxin functioned as an electron carrier. (Received November 4, 1980; Accepted January 9, 1981)     

Bulk-phase viscoelastic properties of seawater relationship with plankton components

The viscous and elastic moduli at different shear rates, togetherwith various biological oceanographic properties, were determinedin seawater from different hydrological layers in the southernNorth Sea in June. The biological oceanographic parameters includedPhaeocystis and Noctiluca abundances, chlorophyll a level (Chl),bacteria. HNAN and aggregate volume fraction. The plankton wasjointly dominated by Phaeocyslis sp. and Noctiluca scinullans.Noctiluca abundance showed no correlation with any other biologicalor viscoelastic parameter, but Phaeocystis abundance correlatedstrongly. The other biological parameters correlated with Phaeocystisand with each other positively and mostly significantly. Overall,viscoelasticity correlated more strongly with Chl than withany other biological parameter. For non-microlayer samples,the excess complex (viscoelastic) modulus (µ.Pa) G^*_E =2.0 x Ch1^1–3 (Chl in mg m^–3). Viscous and elasticmoduli also correlated closely with each other. For a givenvalue of Chl. the microlayer samples were 6.5 or 14 times (dependingon the estimation method) more viscoelastic than in bulk-phasesamples. Viscoelasticity in samples of settled benthic ‘fluff’were lower even than bulk-phase samples, but this differencewas not significant. Comparison with Mediterranean data on viscoelasticity(Jenkinson. Oceanol. Acta, 16, 317–334, 1993), using publishedvalues for phytoplankton biomass (Wiadnyana, J. Rech. Océanogr.,17, 1–6, 1992), suggests that the relationship betweenChl (or phytoplankton biomass) and viscoelasticity might begeneral. This apparent biomodification of the viscosity andelasticity of seawater is discussed in relation to its likelyimpact on turbulence and plankton ecology.     

From pull-down data to protein interaction networks and complexes with biological relevance

Motivation: Recent improvements in high-throughput Mass Spectrometry(MS) technology have expedited genome-wide discovery of protein–proteininteractions by providing a capability of detecting proteincomplexes in a physiological setting. Computational inferenceof protein interaction networks and protein complexes from MSdata are challenging. Advances are required in developing robustand seamlessly integrated procedures for assessment of protein–proteininteraction affinities, mathematical representation of proteininteraction networks, discovery of protein complexes and evaluationof their biological relevance. Results: A multi-step but easy-to-follow framework for identifyingprotein complexes from MS pull-down data is introduced. It assessesinteraction affinity between two proteins based on similarityof their co-purification patterns derived from MS data. It constructsa protein interaction network by adopting a knowledge-guidedthreshold selection method. Based on the network, it identifiesprotein complexes and infers their core components using a graph-theoreticalapproach. It deploys a statistical evaluation procedure to assessbiological relevance of each found complex. On Saccharomycescerevisiae pull-down data, the framework outperformed othermore complicated schemes by at least 10% in F₁-measure and identified610 protein complexes with high-functional homogeneity basedon the enrichment in Gene Ontology (GO) annotation. Manual examinationof the complexes brought forward the hypotheses on cause offalse identifications. Namely, co-purification of differentprotein complexes as mediated by a common non-protein molecule,such as DNA, might be a source of false positives. Protein identificationbias in pull-down technology, such as the hydrophilic bias couldresult in false negatives. Contact: samatovan{at}ornl.gov Supplementary information: Supplementary data are availableat Bioinformatics online. Associate Editor: Jonathan Wren Present address: Department of Biomedical Informatics, VanderbiltUniversity, Nashville, TN 37232. The authors wish it to be known that, in their opinion, thefirst two authors should be regarded as joint First Authors.     

Fitting a geometric graph to a protein-protein interaction network

Motivation: Finding a good network null model for protein–proteininteraction (PPI) networks is a fundamental issue. Such a modelwould provide insights into the interplay between network structureand biological function as well as into evolution. Also, network(graph) models are used to guide biological experiments anddiscover new biological features. It has been proposed thatgeometric random graphs are a good model for PPI networks. Ina geometric random graph, nodes correspond to uniformly randomlydistributed points in a metric space and edges (links) existbetween pairs of nodes for which the corresponding points inthe metric space are close enough according to some distancenorm. Computational experiments have revealed close matchesbetween key topological properties of PPI networks and geometricrandom graph models. In this work, we push the comparison furtherby exploiting the fact that the geometric property can be testedfor directly. To this end, we develop an algorithm that takesPPI interaction data and embeds proteins into a low-dimensionalEuclidean space, under the premise that connectivity informationcorresponds to Euclidean proximity, as in geometric-random graphs.We judge the sensitivity and specificity of the fit by computingthe area under the Receiver Operator Characteristic (ROC) curve.The network embedding algorithm is based on multi-dimensionalscaling, with the square root of the path length in a networkplaying the role of the Euclidean distance in the Euclideanspace. The algorithm exploits sparsity for computational efficiency,and requires only a few sparse matrix multiplications, givinga complexity of O(N²) where N is the number of proteins. Results: The algorithm has been verified in the sense that itsuccessfully rediscovers the geometric structure in artificiallyconstructed geometric networks, even when noise is added byre-wiring some links. Applying the algorithm to 19 publiclyavailable PPI networks of various organisms indicated that:(a) geometric effects are present and (b) two-dimensional Euclideanspace is generally as effective as higher dimensional Euclideanspace for explaining the connectivity. Testing on a high-confidenceyeast data set produced a very strong indication of geometricstructure (area under the ROC curve of 0.89), with this networkbeing essentially indistinguishable from a noisy geometric network.Overall, the results add support to the hypothesis that PPInetworks have a geometric structure. Availability: MATLAB code implementing the algorithm is availableupon request. Contact: natasha{at}ics.uci.edu Associate Editor: Olga Troyanskaya     

Simultaneous measurements of small-scale physical dynamics and zooplankton distributions

Haury et al., (Deep Sea Res., 37, 447–461, 1990) describedhow turbulence and vertical shear generated by the passage ofa storm affected the vertical distributions and community structureof plankton in Monterey Bay, CA. The data presented came froma larger set of five plankton recorder collections of zooplankton,taken together with temperature, salinity, turbulent kineticenergy dissipation rate, and vertical current shear data onthe research submarine Dolphin. Here we report on the analysisof the whole data set, which represents a wide range of physicaldynamic conditions. While the overall community structure remainedrelatively constant over the sampling period, the relationshipsbetween taxonomic categories and between categories and thephysical dynamics were complex and variable; factors relatedto depth were the most important. During periods of weak winds(low kinetic energy conditions) some of the biological variabilityappeared to be related to the level of turbulent dissipation.Other relationships were not detected due to the limitationsof the plankton sampler and its interactions with the highlyvariable biological and physical conditions. In view of thesedifficulties, we recommend that future field investigationsof the relationships between physical parameters and zooplanktondistributions make greater use of acoustic and other remote-sensingtechnologies, preferably in combination with numerical simulationstudies to help guide and interpret the field work.     

A Comparison of Anther Tissue Development in Male Sterile Aloe vera and Male Fertile Aloe ciliaris

Cytological differences between the anther development of amale sterile and a male fertile Aloe species are used to explaininteractions between anther tissues. Some deviations in thelayers of the locule wall and the microspores of the male sterileanther are related to each other and their biological functionsare discussed. The cytological development of the male sterility,which can be observed shortly after meiosis, seems to be restrictedto the locular cavity. The tapetal development and breakdownare normal, apart from the size of some orbicules. However,the pollenkitt is not transported to the pollengrains, whichstrongly supports our theory that this process is mechanicallypollen-controlled. The development of the epidermal and endothecialcells is normal, except in a part of the anthers where thesecells do not expand, after which dehiscence is incomplete. Thelatter process is discussed in relation to the deviations insidethe locular cavity. Aloe vera (L.) Burm. fil., Aloe ciliaris Haw., Liliaceae, male sterility, tapetum, pollenkitt, endothecium, anther dehiscence     

Lateral gene transfer of a multigene region from cyanobacteria to dinoflagellates resulting in a novel plastid-targeted fusion protein

The number of cases of lateral or horizontal gene transfer ineukaryotic genomes is growing steadily, but in most cases, neitherthe donor nor the recipient is known, and the biological implicationsof the transfer are not clear. We describe a relatively well-definedcase of transfer from a cyanobacterial source to an ancestorof dinoflagellates that diverged before Oxyrrhis but after Perkinsus.This case is also exceptional in that 2 adjacent genes, a paralogueof the shikimate biosynthetic enzyme AroB and an O-methyltransferase(OMT) were transferred together and formed a fusion proteinthat was subsequently targeted to the dinoflagellate plastid.Moreover, this fusion subsequently reverted to 2 individualgenes in the genus Karlodinium, but both proteins maintainedplastid localization with the OMT moiety acquiring its own plastid-targetingpeptide. The presence of shikimate biosynthetic enzymes in theplastid is not unprecedented as this is a plastid-based pathwayin many eukaryotes, but this species of OMT has not been associatedwith the plastid previously. It appears that the OMT activitywas drawn into the plastid simply by virtue of its attachmentto the AroB paralogue resulting from their cotransfer and oncein the plastid performed some essential function so that itremained plastid targeted after it separated from AroB. Genefusion events are considered rare and likely stable, and suchan event has recently been used to argue for a root of the eukaryotictree. Our data, however, show that exact reversals of fusionevents do take place, and hence gene fusion data are difficultto interpret without knowledge of the phylogeny of the organisms—thereforetheir use as phylogenetic markers must be considered carefully.     

Multiple imputation when records used for imputation are not used or disseminated for analysis

When some of the records used to estimate the imputation modelsin multiple imputation are not used or available for analysis,the usual multiple imputation variance estimator has positivebias. We present an alternative approach that enables unbiasedestimation of variances and, hence, calibrated inferences insuch contexts. First, using all records, the imputer samplesm values of the parameters of the imputation model. Second,for each parameter draw, the imputer simulates the missing valuesfor all records n times. From these mn completed datasets, theimputer can analyse or disseminate the appropriate subset ofrecords. We develop methods for interval estimation and significancetesting for this approach. Methods are presented in the contextof multiple imputation for measurement error.