Carla Valéria dos Santos Pacheco
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Carla Valéria dos Santos Pacheco has a total of 3 co-authors. Here are the top co-authors by number of shared publications:
- Diego Corrêa Alves
Universidade Estadual de Maringá
1 shared publication - Robson Marcelo Rossi
Universidade Estadual de Maringá, Brazil
1 shared publication - Rodolfo de Souza Nascimento
Universidade Estadual de Maringá
1 shared publication
Latest Publications
- Open Access
- 21 October 2025
- Published by Universidade Estadual de Maringa in Acta Scientiarum. Agronomy
- Vol. 48 (1) , e78972
Bayesian modeling for binomial dose-response using different link functions in bioassays for the control of Diaphorina citri
. This study analyzes the susceptibility of Diaphorina citri, a pest of great relevance to citrus farming, to different concentrations of the entomopathogenic fungi Cordyceps fumosorosea and Beauveria bassiana, which are widely used in biological control. The data were evaluated by means of binomial regression , the linear predictor) with a Bayesian approach, considering different link functions: Basic (logit, probit, cauchit, cloglog, and loglog), power, and reverse power. Model selection was based on the lowest deviance information criterion (DIC) value. For both fungi, the selected models showed a good fit, as indicated by the simulated envelopes of the normalized quantile residuals. For C. fumosorosea, the model with the cauchit link function presented the best fit (DIC = 53.41), and , and a 90% lethal dose (LD90) of 6.38 log(conidia mL-1) [5.83; 7.17]. For B. bassiana, the reverse power cloglog function showed the best performance (DIC = 62.17), and , and an LD90 of 10.69 log(conidia mL-1) . In both models, the regression parameters were statistically significant, while the flexibility parameter (λ) did not show sufficient evidence to justify the use of extended functions. In addition, C. fumosorosea was, in general, more effective than B. bassiana, except in the lower portion of the experimental dose range (4.00 to 4.52 log(conidia mL⁻¹)), in which the latter fungus showed slight superiority. The analysis highlights the importance of a careful choice of the link function in the modeling of binary bioassays, especially regarding decisions on the application of bioinsecticides in agricultural contexts.
- 21 October 2025
- Published by Universidade Estadual de Maringa in Acta Scientiarum. Agronomy
- Vol. 48 (1) , e78972
Open AccessBayesian modeling for binomial dose-response using different link functions in bioassays for the control of Diaphorina citri
. This study analyzes the susceptibility of Diaphorina citri, a pest of great relevance to citrus farming, to different concentrations of the entomopathogenic fungi Cordyceps fumosorosea and Beauveria bassiana, which are widely used in biological control. The data were evaluated by means of binomial regression , the linear predictor) with a Bayesian approach, considering different link functions: Basic (logit, probit, cauchit, cloglog, and loglog), power, and reverse power. Model selection was based on the lowest deviance information criterion (DIC) value. For both fungi, the selected models showed a good fit, as indicated by the simulated envelopes of the normalized quantile residuals. For C. fumosorosea, the model with the cauchit link function presented the best fit (DIC = 53.41), and , and a 90% lethal dose (LD90) of 6.38 log(conidia mL-1) [5.83; 7.17]. For B. bassiana, the reverse power cloglog function showed the best performance (DIC = 62.17), and , and an LD90 of 10.69 log(conidia mL-1) . In both models, the regression parameters were statistically significant, while the flexibility parameter (λ) did not show sufficient evidence to justify the use of extended functions. In addition, C. fumosorosea was, in general, more effective than B. bassiana, except in the lower portion of the experimental dose range (4.00 to 4.52 log(conidia mL⁻¹)), in which the latter fungus showed slight superiority. The analysis highlights the importance of a careful choice of the link function in the modeling of binary bioassays, especially regarding decisions on the application of bioinsecticides in agricultural contexts.