Strategies to develop radiomics and machine learning models for lung cancer stage and histology prediction using small data samples

Predictive models based on radiomics and machine-learning (ML) need large and annotated datasets for training, often difficult to collect. We designed an operative pipeline for model training to exploit data already available to the scientific community. The aim of this work was to explore the capability of radiomic features in predicting tumor histology and stage in patients with non-small cell lung cancer (NSCLC).We analyzed the radiotherapy planning thoracic CT scans of a proprietary sample of 47 subjects (L-RT) and integrated this dataset with a publicly available set of 130 patients from the MAASTRO NSCLC collection (Lung1). We implemented intra- and inter-sample cross-validation strategies (CV) for evaluating the ML predictive model performances with not so large datasets.We carried out two classification tasks: histology classification (3 classes) and overall stage classification (two classes: stage I and II). In the first task, the best performance was obtained by a Random Forest classifier, once the analysis has been restricted to stage I and II tumors of the Lung1 and L-RT merged dataset (AUC = 0.72 ± 0.11). For the overall stage classification, the best results were obtained when training on Lung1 and testing of L-RT dataset (AUC = 0.72 ± 0.04 for Random Forest and AUC = 0.84 ± 0.03 for linear-kernel Support Vector Machine).According to the classification task to be accomplished and to the heterogeneity of the available dataset(s), different CV strategies have to be explored and compared to make a robust assessment of the potential of a predictive model based on radiomics and ML.     

藏药处方使用诃子乌头配伍的概况及减毒机理分析

目的：对藏药处方中诃子与乌头的配伍概况及诃子配伍乌头的减毒机理进行分析,探讨其配伍的机理和意义。方法：从《四部医典》和《藏医秘方精选》中遴选出诃子配伍乌头的相关处方进行分析;通过检索中国期刊全文数据库,查找与诃子及乌头的相关化学成分研究文献,进行遴选分析。结果：从选出相关藏药处方中发现,诃子与乌头配伍,诃子通常用量大于乌头;通过相关文献分析,诃子配伍乌头减毒存在物质基础。结论：藏药处方中乌头常配伍诃子共用,诃子配伍乌头减毒的情况及机理已有初步的分析结果,进一步的相关研究需要继续进行。     

In Vitro Susceptibility of <Emphasis Type="Italic">C. albicans</Emphasis> and <Emphasis Type="Italic">C. neoformens</Emphasis> to Potential Metabolites from <Emphasis Type="Italic">Streptomycetes</Emphasis>

Forty two Streptomycetes isolates from soils of Kodachadri region in Western ghats were recovered by soil dilution technique. Cross streak method was followed for primary screening of antifungal activity. Positive isolates were subjected to secondary screening by cold extraction of fermentation broth in butanol solvent. Six isolates exhibited broad spectrum antifungal activity against all the tested yeast pathogens like Candida albicans, Candida lipolytica, Cryptococcus neoformens and Saccharomyces cerevisiae. One isolate showed excellent antifungal activity against all test organisms with maximum zone of inhibition 60 mm each incase of C. neoformens and C. albicans. Partial characterization of antifungal metabolite by TLC resulted in a purple spot with an R_f value 0.50. The UV absorption spectra at 218 nm indicated possible chemical nature of the active metabolite as polyene group and purity was assessed by analytical HPLC.     

Effect of Neem and Selected Fungicides on Viability and Virulence of the Entomopathogenic Nematode Steinernema feltiae

Entomopathogenic nematodes are often used in conjunction with other pest management tactics and the lack of compatibility information is a major impediment in further expansion of their use. We evaluated the effects of different formulations of neem and selected fungicides commonly used in greenhouses on Steinernema feltiae which is used for the control of fungus gnats. Neem as pure oil at the field recommended concentrations (5- 10 mL L -1 ) had no effect on the viability and virulence of S. feltiae up to 120 h incubation. However the neem formulation, Nimbecidine and neem oil when mixed with a bactericidal soap (commonly used as a surfactant with neem oil) caused 13- 25% mortality of S. feltiae. This toxic effect was entirely due to the soap that alone caused about 24% mortality. Neither neem oil, Nimbecidine or soap had any effect on nematode virulence. The fungicide cinnamaldehyde (Cinnamate) was highly toxic, resulting in 100% nematode mortality after 4 h of incubation, followed by hydrogen dioxide/peroxyacetic acid mixture (ZeroTol) that caused 100% mortality after 120 h of incubation. Another fungicide, azoxystrobin (Abound) caused no nematode mortality. This investigation concludes that neem and the fungicide azoxystrobin (Abound) can be safely tank mixed at the field recommended concentrations with the infective juveniles of S. feltiae for application, but cinnamaldehyde (Cinnamate) and hydrogen dioxide/peroxyacetic mixture (ZeroTol), are incompatible. Also the surfactants that are usually recommended as 'tank-mix' applications can be toxic to the nematodes and should therefore be evaluated for compatibility prior to use.     

An In Vitro Inhibition Test That Predicts Toxicity of Bacterial Pathogen Combinations in the Colorado Potato Beetle

Many insect bacterial pathogens are not toxic enough for field control. Combinations of bacteria may increase toxicity. Bacteria toxic to Colorado potato beetle , Photorhabdus luminescens, Chromobacterium violaceum and Serratia marcescens , were tested in pair-wise combinations in an in vitro double streak test to determine bacterial compatibility. Only C. violaceum and S. marcescens grew to confluency. Their combined toxicity in vivo was additive. Other bacterial combinations had clear zones between bacterial streaks indicating inhibition. In the insect, the combined toxicity was less than the most toxic bacteria of the pair. For these strains, this in vitro test predicted compatibility in the insect.     

Three species groups in American trilliums as revealed by compatibility with an Asian species

American trilliums are all diploid and are classified into three groups with regard to compatibility withTrillium kamtschaticum, a single diploid Asian species: (1) Compatible pedicellate-flowered species, i.e.T. erectum, T. gleasoni, andT. vaseyi. (2) Incompatible pedicellate-flowered species, i.e.T. grandiflorum. (3) Incompatible sessile-flowered species, i.e.T. decunbens, T. luteum, T. sessile andT. stamineum. HybridsT. kamtschaticum×T. erectum flowered for the first time 9–10 years after sowing F₁ seeds. The F₁ were completely sterile in pollen and in seed-setting.     

A four-state cross bridge model for muscle contraction. Mathematical study and validation

A mathematical model of contracting skeletal muscle is studied. The model is composed by an elastic element (SE) in series with a contractile element (CE) that describes the cross bridge kinetics with a formulation derived by that proposed by Eisenberg and Hill (1978). An analytical study of the system of nonlinear partial differential equations of the model allows the existence and the uniqueness of the solution to be proved. A suitable approach to the numerical solution is defined and a series of numerical tests are performed. These tests lead to select an appropriate set of parameters and allow to compare model predictions and experimental observations on frog skeletal muscle. Work partially supported by M.P.I., by CNR through I.A.N. of Pavia and by HUSPI Project     

THE EVOLVABILITY OF GROWTH FORM IN A CLONAL SEAWEED

Although modular construction is considered the key to adaptive growth or growth‐form plasticity in sessile taxa (e.g., plants, seaweeds and colonial invertebrates), the serial expression of genes in morphogenesis may compromise its evolutionary potential if growth forms emerge as integrated wholes from module iteration. To explore the evolvability of growth form in the red seaweed, Asparagopsis armata, we estimated genetic variances, covariances, and cross‐environment correlations for principal components of growth‐form variation in contrasting light environments. We compared variance–covariance matrices across environments to test environmental effects on heritable variation and examined the potential for evolutionary change in the direction of plastic responses to light. Our results suggest that growth form in Asparagopsis may constitute only a single genetic entity whose plasticity affords only limited evolutionary potential. We argue that morphological integration arising from modular construction may constrain the evolvability of growth form in Asparagopsis, emphasizing the critical distinction between genetic and morphological modularity in this and other modular taxa.     

Dehydration-induced cross tolerance of Belgica antarctica larvae to cold and heat is facilitated by trehalose accumulation

Larvae of the Antarctic midge, Belgica antarctica (Diptera: Chironomidae), are frequently exposed to dehydrating conditions on the Antarctic Peninsula. In this study, we examined how rates and levels of dehydration alter heat and cold tolerance and how these relate to levels of trehalose within the insect. When dehydrated, larvae tolerated cold and heat stress more effectively, although resistance to cold was more pronounced than heat resistance. Slow dehydration was more effective than rapid dehydration in increasing temperature tolerance. Severe dehydration (50% reduction in water content) caused a much greater increase in temperature tolerance than did mild dehydration (e.g. 10% water loss). Larvae severely dehydrated at a slow rate (98% RH) were more temperature tolerant than those dehydrated quickly (0 or 75% RH). These results indicate that the slower dehydration rate allows the larvae to more effectively respond to reduced water levels and that physiological adjustments to desiccation provide cross tolerance to cold and heat. Levels of trehalose increased during dehydration and are likely a major factor increasing subsequent cold and heat resistance. This hypothesis was also supported by experimental results showing that injection of trehalose enhanced resistance to temperature stress and dehydration. We conclude that changes in temperature tolerance in B. antarctica are linked to the rate and severity of dehydration and that trehalose elevation is a probable mechanism enhancing this form of cross tolerance.     

Nesting Behavior of <Emphasis Type="Italic">Gorilla gorilla diehli</Emphasis> at Kagwene Mountain,Cameroon: Implications for Assessing Group Size and Density

We recorded nesting data at 569 fresh night nest sites, comprising 7032 individual nests, of Cross River gorillas inhabiting the Kagwene Mountain in western Cameroon. The mean night nest group size was 12.4. Overall, 55% of night nests were constructed on the ground and 45% in trees. Significantly more arboreal nests were constructed in the wet season (69%), vs. the dry season (19%). Day nest construction was common at Kagwene (n = 260 nest sites, mean nest group size = 5.98) and we encountered significantly more day nest sites in the wet season. Nest site reuse was also common (35%), though not related to season. Our results of nesting habits concur with those from other western gorilla studies, in which rainfall influences arboreal nesting. However, we encountered wet season arboreal nesting, day nest construction, and overall nest site reuse more frequently than reported for other sites. Our observations have considerable implications when estimating group size and density using traditional nest count data. The gorillas at Kagwene inhabit the highest altitudinal range of all Cross River gorilla subpopulations and rainfall is also high; therefore other subpopulations may demonstrate different nesting characteristics. However, one should consider our findings when attempting to estimate Cross River gorilla density at other localities through nest site data.