Machine learning highlights the importance of primary and secondary production in determining habitat for marine fish and macroinvertebrates

1. Species distribution models for marine organisms are increasingly used for a range of applications, including spatial planning, conservation, and fisheries management. These models have been constructed using a variety of mathematical forms and drawing on both physical and biological independent variables; however, what might be called first-generation models have mainly followed the form of linear models, or smoothing splines, informed by data collected in the context of fish surveys.
2. The performance of different classes of variables were tested in a series of species occurrence models built with machine learning methods, specifically evaluating the potential contribution of lower trophic level data. Random forest models were fitted based on the classification of the absence/presence for fish and macroinvertebrates surveyed on the US Northeast Continental Shelf.
3. The potential variables included physical, primary production, secondary production, and terrain variables. For accepted model fits, six variable importance measures were computed, which collectively showed that physical and secondary production variables make the greatest contribution across all models. In contrast, terrain variables made the least contribution to these models.
4. Multivariable analyses that account for all performance measures reinforce the role of water depth and temperature in defining species presence and absence; however, chlorophyll concentration and some specific zooplankton taxa, such as Metridia lucens and Paracalanus parvus, also make important contributions with strong seasonal variations.
5. Our results suggest that lower trophic level variables, if available, are valuable in the creation of species distribution models for marine organisms.

     

Ultrastructure of culture forms of the eel trypanosome,Trypanosoma granulosum Laveran & Mesnil, 1902, exposed to polyamine biosynthesis inhibitors

Polyamines are important in the growth, division and differentiation of many cell types, including trypanosomes. The ultrastructure of culture forms of Trypanosoma granulosum was examined following growth in a modified semidefined medium (controls) and in the same medium with added polyamine biosynthesis inhibitors (DFMO, MGBG and Berenil). Untreated trypanosomes had ultrastructural features in common with other cultured flagellates. Those treated with DFMO were generally more rounded in contour. Cytoplasmic vacuoles and swollen mitochondrial membranes suggested osmotic imbalance, perhaps resulting from compromised membrane integrity. An anticlockwise arm dividing subtubule B of the peripheral doublets of the flagellum was noted in some trypanosomes treated with 20 mM DFMO. Vacuolation was also induced by MGBG and flagellar division occurred more frequently at 0.2 mM than in controls. With 1 mM MGBG, mitochondria were difficult to discern and kinetoplasts were disrupted. With 0.2 mM Berenil, vacuoles, swollen mitochondria, membranous whorls, additional microtubules underlying subpellicular tubules and disaggregated kinetoplasts were noted. With 1 mM Berenil, much of the cell structure was destroyed, although the pellicle, flagellum and paraxial rod remained intact. The present study illustrates the importance of hnctional polyamine synthetic pathways for the integrity of membranes, mitochondria, kinetoplasts and possibly microtubules in cultured T. granulosum.     

Influence of multiple amino acid supplementation on the performance of rainbow trout, Oncorhynchus mykiss (Walbaum), fed soya based diets

Using rainbow trout, Oncorhynchus mykiss (Walbaum), of approximately 50 g, a 63 day feeding trial was undertaken to evaluate the efficacy of individual and multiple amino acid supplements in diets where soyabean meal was used as the principle protein source. Iso-nitrogenous diets (45% crude protein) were formulated where the control diet (F) contained fishmeal as the reference protein and a solvent extracted soyabean meal (S) replaced approximately 66% of this protein source. The soyabean containing diets were then supplemented with crystalline amino acids thus; methionine only, dual supplemented with two methionine and lysine levels and finally, a supplement comprising methionine, lysine, tryptophan, threonine, arginine and histidine. The results showed that soyabean meal (SBM) was inferior to the reference protein when SBM was used to replace 66% of the fishmeal and that no significant restoration in growth, feed efficiency and apparent net protein utilization was obtained by either methionine only or dual methionine and lysine supplementation. However, by comparison with the fish fed the unsupplemented, single and dual supplemented diets, multiple amino acid incorporation was associated with significantly improved percentage weight gain, specific growth rate and marginal improvements in apparent net protein utilization. However, performance was not equal to that of the fish fed the fishmeal based control diet. The results are discussed with respect to the level of each of the essential amino acids (EAA) as a proportion of the total EAAs (A/E index) of test diets by reference to the whole body tissue amino acid profile of rainbow trout.     

A serial cross-sectional study of the prevalence of rabbit haemorrhagic disease on three farms in the lower North Island of New Zealand

AIM: To estimate over a 3-year period following the first release of rabbit haemorrhagic disease virus (RHDV) the prevalence of rabbit haemorrhagic disease (RHD) and the abundance of rabbits (Oryctolagus cuniculus) in an area that historically had low rabbit densities.

METHODS: Three farms grazing predominantly sheep and beef cattle, located close together and with low initial rabbit densities, were selected for study. RHDV had been deliberately released on all farms in December 1997. Farms were visited 2–3 times per year between June 1998 and April 2001. At each visit, rabbits were shot with the aid of spotlights at night and blood samples were collected for detection of RHDV antibodies. Rabbit carcasses were necropsied and the age of the animals was determined. Rabbit abundance on each property was measured throughout the study using spotlight night counts. Logistic regression was used to identify factors associated with the risk of carcasses being seropositive for RHDV.

RESULTS: Rabbit density differed initially between farms (8.2, 9.9, 2.3 rabbits per spotlight km in June 1998), and declined on all three properties over time (1.2, 2.4, 1.1 rabbits per spotlight km in November 2000). Highest antibody titres to RHDV were initially evident on the farm on which rabbits were most abundant. The average prevalence of seropositive rabbits overall was 21% (95% CI=15–28%). Female rabbits tended to be less likely to be seropositive for RHDV than males (OR=0.47; 95% CI=0.21–1.02). The odds of becoming seropositive were reduced for rabbits born in the breeding season of 1999–2000 (OR=0.17; 95% CI=0.05–0.64).

CONCLUSIONS: The temporal pattern of outbreaks measured by peaks of seroprevalence differed between closely-spaced farms when they had different rabbit densities, but were similar when rabbit densities were similar. Microclimate and vegetation influencing abundance of insect vectors for RHDV and intrinsic population-related factors like rabbit breeding behaviour are also likely to be involved in local patterns of spread.     

Detection of calves persistently infected with bovine pestivirus in a sample of dairy calves in south-eastern Queensland

Objective To determine the proportion and incidence of calves persistently infected with bovine pestivirus in calves (n = 1521) supplied to the Tick Fever Research Centre and to assess the test regime to detect calves persistently infected with bovine pestivirus.
Design Calves, 1 to 6 weeks old, selected for use in the production of the tick fever vaccine were collected from 21 properties in 56 separate groups between October 1990 and December 1996. Each group was examined for the presence of calves persistently infected with bovine pestivirus.
Procedure All calves were routinely tested for antibody to bovine pestivirus and bovine pestivirus antigen using a serum neutralisation test and an antigen-capture ELISA, respectively. Pooled lymphocyte samples from calves were also monitored for bovine pestivirus by inoculation of sheep. Whole herd testing was carried out in eight herds, using a serum neutralisation test as a screen test followed by an antigen-capture ELISA of cattle with a serum neutralisation test titre of less than 32.
Results Fourteen of the 1521 calves tested (0.9%), were detected as persistently infected and the incidence ranged from 0.0 to 3.0 % per year over 6 years. Persistently infected calves were found in 13 of the 59 groups and originated from 7 of the 21 herds used. In whole herd testing on the properties of origin, cattle persistently infected with bovine pestivirus were detected in four of the eight herds tested
Conclusions The proportion of calves persistently infected with bovine pestivirus is similar to that in other countries and indicates that bovine pestivirus could be a significant cause of economic loss in Australian cattle herds. In detecting calves persistently infected with bovine pestivirus, the combination of sheep inoculation, paired antigen-capture ELISA and serum neutralisation tests appeared to be highly sensitive and specific.     

Multi-centre assessment of mycotic rhinosinusitis in dogs: a retrospective study of initial treatment success (1998 to 2008)

Objectives : To retrospectively review the first treatment response of dogs with mycotic rhinosinusitis to commonly utilised treatment techniques. Methods : Medical records of dogs treated for mycotic rhinosinusitis were obtained retrospectively via a manual review of the clinical databases of six veterinary referral centres for the period of January 1998 to June 2008, and first treatment outcome was evaluated. Historical and clinicopathological findings were also reviewed to evaluate their impact on treatment success or failure. Results : There was no significant difference in first treatment outcome between treatment groups (P=0·21). When all topical treatments were considered together (n=85), 39 dogs (45·8%) had a successful first treatment. Initial treatment success was associated with a younger age (56·3 versus 75·8 months; P=0·02) and was 2·7 times more likely in dogs with unilateral disease, although this was not significant (P=0·07). Adjunctive therapy with systemic antifungal agents was associated with treatment failure (PÄ0·01). Fifty-nine dogs (69·4%) responded successfully following multiple treatments. Clinical Significance : Treatment of mycotic rhinosinusitis remains challenging, and multiple treatments are frequently required for adequate treatment. Reasons for first treatment failure are likely multifactorial in origin, making it difficult to predict those dogs that are likely to have a superior prognosis, regardless of the treatment type used.     

血清中不同甘露(聚)糖结合凝集素浓度的绵羊感染绵羊肺炎支原体后免疫因子表达的变化

为研究中国美利奴羊不同甘露(聚)糖结合凝集素(MBL)浓度感染绵羊肺炎支原体(MO)的免疫因子水平变化,本研究选择血清中MBL高、低浓度的绵羊各6只(感染组和对照组各3只),感染组人工感染MO,分别在人工感染前和感染后不同时间,采用荧光定量PCR法检测血液中血清因子及补体表达水平。结果显示,不同MBL浓度的绵羊人工感染后MBL m RNA水平呈下降趋势,MBL高浓度促炎因子IL-2和IFN-γ的m RNA表达水平较高,抗炎因子IL-4的m RNA表达水平在感染后1 d升高,此后开始下降,而IL-4的m RNA水平在14 d后有所升高;MBL低浓度羊感染后其TNF-α的m RNA水平显著升高,随炎症的缓解,逐渐降低;补体C1和C3的m RNA在感染后表现出不同的变化,MO感染可以激活补体途径。本研究结果表明,低血清MBL浓度与绵羊支原体肺炎具有一定的相关性,MBL不同浓度组之间其IL-2、IL-4、TNF-α、IFN-γ、补体C1、C3水平存在差异,低浓度MBL的绵羊更易发生比较严重的炎症反应。