Assessment of Optimal Selected Prognostic Factors

The identification and assessment of prognostic factors is one of the major tasks in clinical research. The assessment of one single prognostic factor can be done by recently established methods for using optimal cutpoints. Here, we suggest a method to consider an optimal selected prognostic factor from a set of prognostic factors of interest. This can be viewed as a variable selection method and is the underlying decision problem at each node of various tree building algorithms. We propose to use maximally selected statistics where the selection is defined over the set of prognostic factors and over all cutpoints in each prognostic factor. We demonstrate that it is feasible to compute the approximate null distribution. We illustrate the new variable selection test with data of the German Breast Cancer Study Group and of a small study on patients with diffuse large B‐cell lymphoma. Using the null distribution for a p‐value adjusted regression trees algorithm, we adjust for the number of variables analysed at each node as well. (© 2004 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)     

Parasitoid behaviour: predicting field from laboratory

Abstract.  1. Most of what is known about parasitoid behaviour comes from laboratory observations: field quantitative observations on searching parasitoids are extremely difficult to do and are rare. The basic components of Aphytis melinus 's response to California red scale ( Aonidiella aurantii ) were studied in the laboratory: encounter, rejection, drumming, probing, oviposition, and host-feeding. It was then asked whether these observations provided a reliable guide to behaviour in the field in a situation that was very different from the laboratory.
2. Field observations were carried out on bark on the trunk and interior branches of trees where live scale density is extremely high in patches, dead scale make up 90% of all scale, and could be expected to interfere with Aphytis search.
3. The laboratory observations predicted well the time taken in the field for each basic event (drumming or probing) and average times spent on a scale. Also well predicted were the distributions of times spent on drumming, probing, and total time on a scale. Rejection rates were much higher in the field. Thus, the laboratory studies predicted foraging behaviour in the field with variable success; potential explanations for observed mismatch between laboratory and field and its possible larger implications are discussed.     

Prey selection by molluscivorous cichlids foraging on a schistosomiasis vector snail,Biomphalaria glabrata

Summary This paper considers prey size selection by four molluscivorous cichlids feeding on the intermediate host snail of Schistosoma parasites, Biomphalaria glabrata. Haplochromis ishmaeli obtains its prey by crushing the snails between the pharyngeal jaws, whereas H. xenognathus, H. sauvagei and Macropleurodus bicolor apply both pharyngeal crushing and oral shelling. The fishes crushed significantly more snails with the highest reward in biomass per second of crushing. Oral shelling occurred far less often than pharyngeal crushing. Encounter rates with prey showed significant variations between different size classes of prey. The fish have no overall knowledge of snail availability in a tank. The probability that a snail will be eaten at encounter, calculated from the number encountered and the number eaten, reflects the prey size preference of the fish. Those snails with the highest biomass/crushing-time ratio had the highest probability of being crushed; observed and predicted prey size preferences corresponded well. Although for oral shelling the potential reward in biomass per second is of the same magnitude as for crushing, the probability of successful shelling is very low. Apparently the fish prefer prey with lowest risks.     

The time and egg budget of Leptopilina clavipes, a parasitoid of larval Drosophila

Abstract.

• 1 For the understanding of the influence of natural selection on the persistence of host selection behaviour in populations of Drosophila parasitoids it is important to know whether parasitoids will become time- or egg-limited. We investigated whether the Drosophila parasitoid Leptopilina clavipes (Hartig) meets egg- or time-limited conditions in the field.
• 2 To this end the following aspects of the parasitoid's life were studied: egg load at emergence, travelling velocity between patches, patch residence times, oviposition rates and life expectancy. Together with the results from earlier studies on host and patch distributions, this formed the input of a ‘Monte Carlo’ simulation model, in which the life history of an individual parasitoid can be traced.
• 3 The simulations revealed that under the conditions found in the field 12.9% of the parasitoid population is egg-limited. The model was also run for a number of scenarios which reflect ‘good’ or ‘bad’ circumstances. In most cases a significant proportion of the parasitoid population proved to be egg-limited.
• 4 For the measurement of travelling velocities and patch residence times a marking method, especially applicable to small-sized parasitoids such as L.clavipes, is described. Marking did not affect survival, host habitat location or host detection rate. Parasitoids were found to be attracted to the odour of fruit-bodies of Phallus impudicus, the most important breeding substrate of their hosts.

     

Is male back space limiting? An investigation into the reproductive demography of the giant water bug,Abedus indentatus (Heteroptera: Belostomatidae)

Because male giant water bugs in the subfamily Belostomatinae provide parental care by brooding eggs on their back, an accurate assessment can be made of both the actual and potential reproductive capacity of males. Two operational sex-ratio (OSR) indices were developed and empirically measured for a population of giant water bug, Abedus indentatus,in California. One index was based on reproductive rates measured in the laboratory; the other index was based on reproductive resources observed in the field. Both OSR indices suggest that the operational sex ratio fluctuates between maleskewed ratios in the summer and femaleskewed ratios in the winter. This pattern appears to be the consequence of two factors. First, the adult sex ratio is significantly female biased. Second, although males can outreproduce females at high ambient temperatures, the reverse is true at low temperatures. Possible reasons for the female-skewed adult sex ratio are examined, including differential recruitment, differential mortality, and sampling bias.     

Springtail postmolt vulnerability to pseudoscorpion predation: Mechanisms and implications

Arthropod prey are expected to be more vulnerable to their predators immediately following molt. The effects of springtail (Isotoma carpenteri) postmolt vulnerability on interactions with a pseudoscorpion predator were examined in the laboratory. Springtails exposed to vials pretreated with pseudoscorpions (Apochthonius minimus) delayed molting compared to those prey that were exposed to vials pretreated only with springtails. Although their escape ability (measured as distance jumped) was unaffected by molt condition, postmolt springtails were more profitable in terms of reduced predator handling time following capture. Despite this,A. minimus did not distinguish between postmolt and intermolt prey presented at either end of a T-maze.     

The potential for rust infection to cause natural selection in apomictic Arabis holboellii (Brassicaceae)

Few studies have examined the potential for pathogens with complex life cycles to cause selection on their required alternate (=intermediate) hosts. Here we examine the effects of two fungal pathogens on an herbaceous mustard, Arabis holboellii. One pathogen species uses A. holboellii as a primary host, the other uses it as an alternate host. This plant-pathogen system is especially interesting because the host, A. holboellii, is apomictic; thus individuals reproduce exact copies of themselves. Despite this mode of reproduction, A. holboellii populations are surprisingly genetically diverse. Could frequency dependent selection by pathogens be maintaining clonal diversity? This study assesses the potential for selection by pathogens. In a controlled greehouse experiment we show that there is heritable variation in A. holboellii's resistance to the rust, Puccinia monoica, and that host fitness is severely reduced by P. monoica infection in both the greenhouse and under natural conditions. Field observations indicate that host clones are also differentially susceptible to the short-cycled rust, P. thlaspeos, and that host fitness is reduced by infection to this pathogen as well. Although the preconditions for pathogen-mediated selection are present, frequency-dependent selection by pathogens is unlikely to be important in structuring populations of Arabis holboellii because multiple host genotypes are susceptible to the same inoculum and the pathogen has a long generation time.