A comparative study of cell classifiers for image-based high-throughput screening

Background

Millions of cells are present in thousands of images created in high-throughput screening (HTS). Biologists could classify each of these cells into a phenotype by visual inspection. But in the presence of millions of cells this visual classification task becomes infeasible. Biologists train classification models on a few thousand visually classified example cells and iteratively improve the training data by visual inspection of the important misclassified phenotypes. Classification methods differ in performance and performance evaluation time. We present a comparative study of computational performance of gentle boosting, joint boosting CellProfiler Analyst (CPA), support vector machines (linear and radial basis function) and linear discriminant analysis (LDA) on two data sets of HT29 and HeLa cancer cells.

Results

For the HT29 data set we find that gentle boosting, SVM (linear) and SVM (RBF) are close in performance but SVM (linear) is faster than gentle boosting and SVM (RBF). For the HT29 data set the average performance difference between SVM (RBF) and SVM (linear) is 0.42 %. For the HeLa data set we find that SVM (RBF) outperforms other classification methods and is on average 1.41 % better in performance than SVM (linear).

Conclusions

Our study proposes SVM (linear) for iterative improvement of the training data and SVM (RBF) for the final classifier to classify all unlabeled cells in the whole data set.

Electronic supplementary material

The online version of this article (doi:10.1186/1471-2105-15-342) contains supplementary material, which is available to authorized users.     

Environmental flows and the European Water Framework Directive

1. Environmental flows is now a widely accepted term that covers the quantity, timing, duration, frequency and quality of water flows required to sustain freshwater, estuarine and near-shore ecosystems and the human livelihoods and well-being that depend on them.
2. The Water Framework Directive (WFD) of the European Union does not use the term environmental flows explicitly, but requires member states to achieve good ecological status (GES) in all waterbodies, which is assessed by reference to aquatic biology. Nevertheless, it is accepted that ecologically appropriate hydrological regimes are necessary to meet this status. Implementing environmental flows will be a key measure for restoring and managing river ecosystems.
3. The WFD explicitly requires stakeholder involvement, but this has been interpreted as largely a dissemination exercise by national government agencies. Stakeholders are no longer involved in negotiation over ecological objectives as these are pre-set in the WFD. However, stakeholders may be more involved in reviewing standards and agreeing to measures to restore river ecosystems to the status required by the WFD.
4. The U.K. has undertaken two major projects to set environmental standards for water resources (i) to define water abstraction limits that maintain a healthy river ecosystem and (ii) to define ecologically appropriate flow releases from reservoirs.
5. Implementation of environmental flows remains a major issue, but new ideas such as time-limited licences and licence trading are being tried.     

Study of the mitotic and meiotic chromosomes of sotol (<i>Dasylirion cedrosanum</i> Trel.)

The genus Dasylirion is a group of plants typically present in the Chihuahuan Desert, perennial, with a dioecious sexual behavior and commonly called sotoles. This genus has been little studied from the biological point of view, and the bases of its reproductive response remain unknown. In this work we studied the chromosome number and meiotic response of Dasylirion cedrosanum in the county of Saltillo, Coahuila, located at the North East of Mexico. For the preparation of mitotic chromosomes, we used a technique based on enzymatic treatment with pectolyase and cellulase, as well as staining with acetocarmin dye. For the study of meiosis, male flower buds were collected, fixed and stained for analysis with the same dye. As a result, the gametic (n = x = 19) and somatic chromosome (2n = 38) numbers of D. cedrosanum are reported for the first time, being consistent with previous findings in other Dasylirion species, which points to a constant ploidy level across the genus. Variation was observed in the morphology and size of the somatic chromosomes, with types ranging from submetacentric to subtelocentric, and sizes oscillating in a range of 4.43 µm, with an average total length of 112.38 µm for the diploid chromosome complement. This shows that the chromosome complement of D. cedrosanom would belong to a 3B classification of Stebins, with a medium variation between chromosome lengths and low chromosome asymmetry. This variation indicates the feasibility of constructing a chromosome ideotype for this species. The meiotic chromosome pairing showed a chromosome behavior consistent with a disomic inheritance characteristic of a diploid species, with prevalence of ring and chain bivalents, typically without pairing abnormalities. Bivalent configurations in all cases were symmetrical.The normal and symmetrical meiotic pairing indicates a balanced production of gametes, and suggests the absence of heteromorphic sex determination.     

Observations on the Formation and Function of the Root Nodules of Alnus glutinosa (L.) Gaertn

Nodulated alder plants grow vigorously in water culture withoutcombined nitrogen. Evidence is advanced to show that the fixationof atmospheric nitrogen thus implied occurs actually withinthe nodulated plant and probably in the nodule. Nodule formation occurred most freely over the pH range 5•4to 7•0, while subsequent to nodulation the best growthof plants was in the pH range 4•2 to 5•4. The capacityof the host plant to tolerate relatively low pH levels considerablyexceeds that of the nodule organism. The oxygen requirementof the nodules appears to be relatively high. The fixation ofnitrogen per unit dry weight of nodule tissue exceeds that oflegumes grown under comparable conditions.     

High growing temperatures reduce fruit carbohydrate and vitamin C in kiwifruit

Kiwifruit vines are perennial plants grown in climates varying from maritime to continental. To determine key responses to temperature, vines were heated at different stages of fruit development, and vine growth and fruit composition examined. Heating vines during fruit starch accumulation caused a major shift in partitioning towards vegetative growth and dramatically reduced fruit carbohydrate and vitamin C. In the following season, growth and flowering were severely reduced. Heating vines during fruit cell division had minimal long‐term effects, whereas heating during fruit maturation delayed starch degradation and fruit ripening and affected growth in the following season. When vines were removed from heat, fruit dry matter, starch and sugar levels were always reduced but hexose : sucrose ratios and inositol were raised. Heating vines affected expression of two sucrose synthase genes, but this did not correlate with reduction in fruit carbohydrate. Activity and expression of l ‐galactose dehydrogenase decreased as fruit developed, suggesting some vitamin C biosynthesis must take place in the fruit. Activity and expression of actinidin increased in response to heat. The results of this study have demonstrated both short and long‐term plant responses to elevated temperatures in woody perennials, and that the timing of heat exposure has severe consequences for vitamin C levels in fruit.