No abstract

Abstract. This work proposes a classification-rule discovery algorithm integrating artificial immune systems and fuzzy systems. The algorithm consists of two parts: a sequential covering procedure and a rule evolution procedure. Each antibody (candidate solution) corresponds to a classification rule. The classification of new examples (antigens) considers not only the fitness of a fuzzy rule based on the entire training set, but also the affinity between the rule and the new example. This affinity must be greater than a threshold in order for the fuzzy rule to be activated, and it is proposed an adaptive procedure for computing this threshold for each rule. This paper reports results for the proposed algorithm in several data sets. Results are analyzed with respect to both predictive accuracy and rule set simplicity, and are compared with C4.5rules, a very popular data mining algorithm.

Neotropical Entomology 31(1): 091-099 (2002) Efeitos de Diferentes Formulações na Viabilidade e no Armazenamento de Conídios deMetarhizium anisopliae a Médio Prazo ABSTRACT -The aim of this work was to evaluate the effects of different formulations on the viability of Metarhizium anisopliae conidia. In a first experiment, the viability of conidia mixed with eight emulsifiable adjuvant oils (EAO), seven wetter/spreaders, three vegetable oils and four mineral oils was evaluated 24h and 48h after spreading over SDA medium surface. Some formulations, which did not present any adverse effect on conidial viability in the first 24h of incubation, were recommended for the medium-term storage experiment. In a second experiment, the effects of different formulations on medium-term storage (40 weeks) of M. anisopliae var. acridum conidia were evaluated at 10ºC and 27ºC. Five EAO, one vegetable oil, a mixture of mineral oils and pure dry conidia were tested. In the experiment 1, the oil formulations did not cause any negative effect on conidial germination. The treatments with the adjuvant oils Actipron, Ashlade and Codacide gave germination levels equal to mineral and vegetable oils after 24h of incubation. The wetter/spreader Ethoken C12 was lethal to conidia. All tested emulsifiable adjuvant oil-based formulations were compatible to the conidia as the oil-based formulations after 48h of incubation. In the experiment 2, conidial viability within the same formulation declined over time. Conidial viability was maintained above 90% at 10ºC in all tested formulants. The conidial suspension with Ashlade gave equal germination levels to Shellsol plus Ondina, when stored at 10ºC and 27ºC for 40 weeks. Thus EAO can be used to formulate M. anisopliae conidia without adverse effects on viability.

Abstract. This work proposes two versions of an Artificial Immune System (AIS) -a relatively recent computational intelligence paradigm -for predicting protein functions described in the Gene Ontology (GO). The GO has functional classes (GO terms) specified in the form of a directed acyclic graph, which leads to a very challenging multi-label hierarchical classification problem where a protein can be assigned multiple classes (functions, GO terms) across several levels of the GO's term hierarchy. Hence, the proposed approach, called MHC-AIS (Multi-label Hierarchical Classification with an Artificial Immune System), is a sophisticated classification algorithm tailored to both multi-label and hierarchical classification. The first version of the MHC-AIS builds a global classifier to predict all classes in the application domain, whilst the second version builds a local classifier to predict each class. In both versions of the MHC-AIS the classifier is expressed as a set of IF-THEN classification rules, which have the advantage of representing comprehensible knowledge to biologist users. The two MHC-AIS versions are evaluated on a dataset of DNAbinding and ATPase proteins.

Abstract-This work presents a system for knowledge discovery from protein databases, based on an Artificial Immune System. The discovered rules have the advantage of representing comprehensible knowledge to biologist users. This task leads to a very challenging problem since a protein can be assigned multiple classes (functions or Gene Ontology (GO) terms) across several levels of the GO's term hierarchy. To solve this problem we present two versions of an algorithm called MHC-AIS (Multi-label Hierarchical Classification with an Artificial Immune System), which is a sophisticated classification algorithm tailored to both multi-label and hierarchical classification. The first version of MHC-AIS builds a global classifier to predict all classes in the dataset, whilst the second version builds a local classifier to predict each class. The proposed versions and an algorithm chosen for comparison are evaluated on a protein dataset, and the results show that MHC-AIS outperformed the compared algorithm in general.

Microbial pest control agents or entomopathogens have been considered an interesting alternative to use instead of chemical insecticides. Knowledge of ecotoxicity data is very important to predict the hazard of any product released in the environment and subsidize the regulation of these products by governmental agencies. In the present study four new Brazilian strains of Bacillus and one fungus were tested to evaluate their acute toxicity to the microcrustacean Daphnia similis, the snail Biomphalaria glabrata and the dung beetle Digitonthophagus gazella. The microcrustaceans and the snails were exposed to entomopathogens in synthetic softwater and the beetles were exposed directly in cattle dung. Obtained data reveal low susceptibility of the non-target species to tested microorganisms, with lethal concentrations being observed only at much higher concentrations than that effective against target insects. These results show that the tested strains are selective in their action mode and seem to be non-hazardous to non-target species.

No abstract

-The fungus Metarhizium anisopliae var. acridum, strain CG 423, was tested under field conditions against the gregarious grasshopper Rhammatocerus schistocercoides (Rehn) (Orthoptera: Acrididae). Conidia formulated in a racemic mixture of soybean oil and kerosene were sprayed under field conditions using an ultralow-volume hand-held atomizer Ulva Plus adjusted to deliver 2.9 L/ha. Bands composed of 2 nd instar nymphs were treated with either 5.0x10 12 or 1.0x10 13 viable conidia/ha. The number of insects in each band was estimated at day one following spraying and by the end of the field trial (15 to 16 days posttreatment). Reductions in population size reached, in average, 65.8% and 80.4% for bands treated with the higher and lower dosage, respectively. For both dosages, total mortality rates of insects collected at two days post-application, and kept in cages for 14 days under lab conditions, showed no significant differences as compared to that obtained with insects collected immediately after spraying. Healthy insects were fed to native grasses sprayed on the field with 1.0x10 13 viable conidia/ha. Mortality levels of the nymphs fed on grasses collected two and four days post-application were not affected when compared to nymphs fed on grasses collected immediately following application.Index terms: biological control, microbial pesticide, entomogenous fungus, grasshopper, oil formulation. Efeito de duas dosagens de Metarhizium anisopliae var. acridum contra Rhammatocerus schistocercoides RehnResumo -O fungo Metarhizium anisopliae var. acridum, isolado CG 423, foi avaliado em condições de campo como agente de controle biológico do gafanhoto gregário Rhammatocerus schistocercoides (Rehn) (Orthoptera: Acrididae). Conídios formulados em uma mistura racêmica de óleo de soja e querosene foram pulverizados no campo com a utilização de um atomizador rotativo manual Ulva Plus, calibrado para aplicação de 2,9 L/ha. Bandos com ninfas de 2 o estádio foram tratados com o equivalente a 5,0x10 12 ou 1,0x10 13 conídios viáveis/ha. O número de insetos em cada bando foi estimado no 1 o dia após a aplicação e ao final do experimento (15 a 16 dias após tratamento). Reduções populacionais médias de 65,8% e 80,4% foram observadas nos bandos tratados com a maior e menor dosagem, respectivamente. Para as dosagens avaliadas, a mortalidade total de insetos coletados aos dois dias após a aplicação e mantidos em gaiolas durante 14 dias, no laboratório, não apresentou diferença significativa daquela obtida para insetos coletados imediatamente após a pulverização. Insetos sadios foram alimentados com gramíneas nativas pulverizadas no campo com 1,0x10 13 conídios viáveis/ha. As taxas de mortalidade observadas nas ninfas alimentadas com capim coletado aos dois e quatro dias após a pulverização não foram afetadas quando comparadas com ninfas alimentadas com capim coletado imediatamente após a pulverização.Termos para indexação: controle biológico, pesticida microbiano, fungo entomopatogênico, gafanhoto, formulação oleosa.