The use of infrared spectroscopy and machine learning tools for detection of Meloidogyne infestations

Plant parasitic nematodes are generally soilborne pathogens that attack plants and cause economic losses in many crops. The infested plants show nonspecific symptoms or, often, are symptomless; therefore, diagnosis is performed by taking soil and root tissue samples. Here, we show that a combination of different infrared spectra analysis and machine learning algorithms can be used to detect plant parasitic nematode infestations before symptoms become visible, using leaves instead of roots and soil as samples. We found that tomato and guava plants infested with Meloidogyne enterorlobii produced different spectral patterns compared to uninfested plants. Using partial spectra from 1,450 to 900/cm as the "fingerprint region", principal component analyses indicated that after 5 (tomatoes) or 8 weeks (guava), plants with no visible symptoms of infestations were positively diagnosed. To improve the early detection response, we used machine learning modelling. A support vector machine (SVM) was used to obtain more robust, accurate models. The SVM model contained 34 support vectors, 17 for each level. The overall performance of the model was >97% and the total accuracy was significantly higher, demonstrating the absence of chance prediction. The best prediction of infestation was obtained at the second and fourth weeks for tomatoes and guavas, respectively, reducing the diagnostic time by half. The combined application of these techniques reduces the processing time from field to laboratory and shows enormous advantages by avoiding root and soil sampling.     

Applying fire connectivity and centrality measures to mitigate the cheatgrass-fire cycle in the arid West,USA

Context

Strategic placement of fuel treatments across large landscapes is an important step to mitigate the collective effects of fires interacting over broad spatial and temporal extents. On landscapes where highly invasive cheatgrass (Bromus tectorum) is increasing fire activity, such an approach could help maintain landscape resilience.

Objectives

Our objectives are to 1) model and map fire connectivity on a cheatgrass-invaded landscape, as well as the centrality of large cheatgrass patches, in order to inform a landscape fuel treatment (i.e., a network of greenstrips); and 2) evaluate the modeled greenstrip network based on changes to cheatgrass patch centrality.

Methods

Our analysis covers 485-km² on the Kaibab National Forest in Northern Arizona. We apply a circuit-theoretic model of fire connectivity between all pairs of large cheatgrass patches. Based on these results, we calculate a measure of centrality for each patch to inform fuel treatment placement. We evaluate the modeled greenstrip network by comparing the pre- and post-treatment centrality of each patch.

Results

After modeling fire connectivity across the landscape, we identify 25 of 68 large cheatgrass patches with relatively high centrality. When we simulate greenstrips around these focal patches, model results suggest that they are effective in reducing the centrality for at least 19 of the 25 patches.

Conclusions

Fire connectivity models provide robust network centrality measures, which can help generate multiple, landscape fuel treatment alternatives and facilitate on-the-ground decisions. The extension of these methods is well suited for landscape fuels management in other vegetation communities and ecosystems.

     

Histophagous scuticociliatids (Ciliophora) parasitizing turbot Scophthalmus maximus: morphology, in vitro culture and virulence

Systemic ciliatosis caused by histophagous ciliates constitutes a serious disease of cultured turbot. Six ciliate isolates were obtained from parasitized turbot during six epizootics at four different farms located in Spain, France and Portugal. Axenic cultures of the six isolates were obtained by periodical subculturing in ATCC 1651MA or supplemented L-15 media. In basal media or seawater, the parasites could survive starving for long periods with no apparent proliferation. In adequate media, growth kinetics was found to be very similar for isolates A and B, with a clear influence of temperature. Morphological studies demonstrated that all isolates share common features that allows their assignment to either Philasterides Kahl, 1931 or Miamiensis Thompson et Moewus, 1964. However, statistically significant differences were evident in pairwise comparisons of the isolates from the four farm sites in 16 taxonomically relevant morphometric features. This could allow the discrimination of different species or strains. Virulence of isolates A and B for healthy turbot was tested in several experiments. Differences in the virulence were especially evident after long-term in vitro culturing, isolate A being clearly attenuated after 35-42 passages, whereas isolate B became more virulent after 20-42 passages. The need of further studies to confirm such virulence variability and its implications in pathogenesis and prevention of turbot scuticociliatoses is stressed.     

Effect of spineless cactus intake (<Emphasis Type="Italic">Opuntia ficus-indica</Emphasis>) on blood glucose levels in lactating sows and its impact on feed intake,body weight loss,and weaning-estrus interval

The effect of spineless cactus intake (Opuntia ficus-indica) on blood glucose (BG) levels in lactating sows and its impact on daily and total feed intake (dFI^?1 and TFI, respectively), body weight loss (BWL), and weaning-estrus interval length (WEI) were evaluated. Thirty-four hybrid (Yorkshire × Landrace × Pietrain) sows in lactation phase were used. Sows were divided into two groups: G1 (n = 17) where they received commercial feed and G2 (n = 17) provided with commercial feed plus an average of 2.0 ± 0.5 kg spineless cactus, based on a sow’s body weight. The variables evaluated were BG, dFI^?1, TFI, BWL, and WEI. Statistical analysis was performed by using a fixed and mixed model methodology, under a repeated measurements experiment. Group effects were found on all analyzed variables (P < 0.05). The BG was lower in G2 (55.2 and 64.5 mg/dL pre- and post-prandial, respectively), compared to that in G1 (70.9 and 80.1 mg/dL pre- and post-prandial, respectively) (P < 0.05). G2 showed better performance than G1 for dFI^?1, BWL, and WEI (P < 0.05) whose averages were 5.5 ± 1.8 kg, 7.4 ± 4.5%, and 5.3 ± 1.2 days, respectively. Averages for these variables in G1 were 4.7 ± 1.5 kg, 16.8 ± 4.6%, and 6.1 ± 1.6 days, respectively. Intake of spineless cactus reduced BG levels in lactating sows, generating greater dFI^?1, lower BWL at the end of lactation, and a lower WEI.     

C-reactive protein measurement in canine saliva.

An established time-resolved immunofluorometric assay designed for measurement of C-reactive protein (CRP) in canine blood was evaluated and validated for use in canine saliva. C-reactive protein was measured in saliva specimens from 5 healthy dogs before and after the injection of casein and in 37 dogs with different disease conditions. The analytical and functional limits of detection were 0.000053 microg/ml and 0.0091 microg/ml, respectively, and intra- and interassay coefficients of variation ranged between 6.7-9.9% and 8.5-16.5%, respectively. A recovery experiment showed no significant disagreement between detected values and expected ones, and saliva CRP concentration was measured in a linear and proportional manner. A positive correlation was found between CRP levels obtained in saliva and serum samples in the experimental (R2 = 0.76) and clinical studies (R2 = 0.70). The assay was able to detect significant differences between salivary CRP levels in healthy dogs and dogs with inflammatory processes. These results suggest that saliva can be used for CRP measurement in dogs. The use of saliva presents the advantage of an easier and less stressful sampling method for the animals, which might be performed outside of hospital environments.     

Estimation of genetic parameters for growth traits of Japanese flounder <Emphasis Type="Italic">Paralichthys olivaceus</Emphasis> using an animal model

To estimate genetic parameters of growth traits in Japanese flounder Paralichthys olivaceus, full-sib and half-sib families were produced in three consecutive years at the Beidaihe Central Experiment Station in China. Each year 8–28 families were produced. The body weight, body length and body depth at 180, 240, and 360 days of age were measured for 5,224 individuals. Four animal models were used to examine the phenotypic variation of growth traits and were compared using the likelihood ratio test. The results showed that estimates for additive genetic effect heritabilities varied greatly depending on the model, trait and age. The maternal effect had a significant impact on phenotypic variation only for body depth at 180 days of age, which explained 49% of the phenotypic variance. The ratio of full-sib effect to phenotypic variation ranged from 0.09 to 0.22. Growth traits all exhibited low heritability (0.13–0.39), indicating that there is the potential for family selection breeding for these traits in Japanese flounder. Using the full model with the fixed, full-sib family, additive and maternal genetic effects, genetic correlations among the three traits for fish of the same age were estimated to be more than 0.80. Generally, the genetic correlations gradually increased as age increased.     

Relationship between the incidence of latent infections caused by <Emphasis Type="Italic">Monilinia</Emphasis> spp<Emphasis Type="Italic">.</Emphasis> and the incidence of brown rot of peach fruit: factors affecting latent infection

Five field experiments were performed in commercial orchards located in Lleida (Spain) over three growing seasons, 2000–2002, in order to estimate the relationship between the incidence of latent infection caused by Monilinia spp. in peaches and the incidence of post-harvest brown rot. No latent infection was recorded at popcorn and the maximum incidence occurred pre-harvest; in some orchards a second peak was detected during the pit hardening period. Monilinia laxa is the most prevalent species isolated from peaches with brown rot. There was a positive correlation between the incidence of latent infection and that of post-harvest brown rot. The average incidence of latent infection during the crop season explained 55% of the total variation in the incidence of post-harvest brown rot. The effect of temperature (T) and duration of wetness (W) on the incidence of latent infection in peach and nectarine orchards was analysed using multiple regression. The regression analysis indicated that T and W jointly explained 83% of the total variation in the incidence of latent infection. The model predicts no latent infections when T < 8°C, and >22 h wetness are required when T = 8°C but only 5 h at 25°C are necessary for latent infection to occur. The incidence of brown rot and latent infection of peaches caused by M. laxa under controlled experimental conditions were also affected by T and W, as well as by fruit maturity and inoculum concentration. Latent infections were produced in fruit when T was not suitable for the development of brown rot symptoms. In these experiments more than 4–5 h of daily wetness were required after embryo growth in fruit sprayed to run-off with an inoculum concentration higher than 10⁴ conidia ml⁻¹ of M. laxa for brown rot and latent infections to develop. The fitted model obtained from the field data was able to predict the observed data obtained under controlled environmental conditions.     

Increase in chemokines CXCL10 and CCL2 in blood from pigs infected with high compared to low virulence African swine fever virus isolates

Modulation of the expression of chemokines and chemokine receptors in whole blood was compared following infection of pigs with high and low virulence isolates of African swine fever virus. Levels of mRNAs for CCL2, CCL3L1, CCL4, CXCL10, CCR1 and CCR5 were significantly increased in at least one time point following infection in two experiments and CCL5, CCR9 and CXCR4 mRNA were significantly increased in one of the experiments. The results showed that greatest fold increases in mRNAs for CXCL10 and CCL2 were observed following infection of pigs. CXCL10 mRNA was increased by up to 15 fold in infected compared to uninfected pigs. CXCL10 protein was also detected in serum from pigs infected with the high virulence Benin 97/1 isolate. Levels of CCL2 mRNA were increased in pigs infected with high virulence Benin 97/1 isolate compared to low virulence OURT88/3 isolate and this correlated with an increase of greater than 30 fold in levels of CCL2 protein detected in serum from pigs infected with this isolate. An increase in overall chemotaxis active compounds in defibrinated plasma samples from Benin 97/1 infected pigs was observed at 3 days post-infection (dpi) and a decrease by 7 dpi as measured by chemotaxis assay using normal pig leucocytes in vitro. Increased levels of CXCL10 may either contribute to the activation of lymphocyte priming toward the Th1 phenotype or induction of T lymphocyte apoptosis. Increased levels of CCL2, a chemoattractant for macrophages, may result in increased recruitment of monocytes from bone marrow thus increasing the pool of cells susceptible to infection.