Gaussian mixtures for intensity modeling of spots in microscopy

Pan, Kangyu; Kokaram, Anil; Hillebrand, Jens; Ramaswami, Mani

doi:10.1109/isbi.2010.5490398

Cited by 6 publications

(9 citation statements)

References 10 publications

(11 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The details of the signal-processing algorithms used have been described elsewhere (Pan et al, 2010). However, the main steps and logic were as follows.…”

Section: Methodsmentioning

confidence: 99%

The Me31B DEAD-Box Helicase Localizes to Postsynaptic Foci and Regulates Expression of a CaMKII Reporter mRNA in Dendrites of Drosophila Olfactory Projection Neurons

Hillebrand¹,

Pan²,

Kokaram³

et al. 2010

Front. Neural Circuits

Self Cite

View full text Add to dashboard Cite

mRNP granules at adult central synapses are postulated to regulate local mRNA translation and synapse plasticity. However, they are very poorly characterized in vivo. Here, in Drosophila olfactory synapses, we present early observations and characterization of candidate synaptic mRNP particles, one of which contains a widely conserved, DEAD-box helicase, Me31B. In Drosophila, Me31B is required for translational repression of maternal and miRNA-target mRNAs. A role in neuronal translational control is primarily suggested by Me31B's localization, in cultured primary neurons, to neuritic mRNP granules that contain: (i) various translational regulators; (ii) CaMKII mRNA; and (iii) several P-body markers including the mRNA hydrolases, Dcp1, and Pcm/Xrn-1. In adult neurons, Me31B localizes to P-body like cytoplasmic foci/particles in neuronal soma. In addition it is present to synaptic foci that may lack RNA degradative enzymes and localize predominantly to dendritic elements of olfactory sensory and projection neurons (PNs). MARCM clones of PNs mutant for Me31B show loss of both Me31B and Dcp1-positive dendritic puncta, suggesting potential interactions between these granule types. In PNs, expression of validated hairpin-RNAi constructs against Me31B causes visible knockdown of endogenous protein, as assessed by the brightness and number of Me31B puncta. Knockdown of Me31B also causes a substantial elevation in observed levels of a translational reporter of CaMKII, a postsynaptic protein whose mRNA has been shown to be localized to PN dendrites and to be translationally regulated, at least in part through the miRNA pathway. Thus, neuronal Me31B is present in dendritic particles in vivo and is required for repression of a translationally regulated synaptic mRNA.

show abstract

“…The details of the signal-processing algorithms used have been described elsewhere (Pan et al, 2010). However, the main steps and logic were as follows.…”

Section: Methodsmentioning

confidence: 99%

The Me31B DEAD-Box Helicase Localizes to Postsynaptic Foci and Regulates Expression of a CaMKII Reporter mRNA in Dendrites of Drosophila Olfactory Projection Neurons

Hillebrand¹,

Pan²,

Kokaram³

et al. 2010

Front. Neural Circuits

Self Cite

View full text Add to dashboard Cite

show abstract

“…As described in the previous work [2], a pre-processing is implemented before the modeling operation. In the first step, the raw image is lightly sharpened and denoised with a Wiener filter using a 7 × 7 disk kernel.…”

Section: Pre-processingmentioning

confidence: 99%

“…However, as shown in Fig.2, the background brightness in the fluorescent image is inhomogeneous. In order to identify the spots without the background bias, the second step exploits a modified contrast method [2]. The spots are detected as regions of high contrast in the image.…”

Section: Pre-processingmentioning

confidence: 99%

“…The fundamental problem is that the positions of the new means µ i and µ j are undefined. Previous work [2] implemented Zhang's method [5] to propose a random sample for the new means. This can lead to an unrepeatable process since the EM algorithm would converge to different solutions on every invocation.…”

Section: Split-stepmentioning

confidence: 99%

“…Moreover, a new split/merge step is applied to the modeling algorithm. In previous work [2], the split-merge algorithm relied on randomly sampled split proposals hence resulting in a non-repeatable process. In our work, we use the estimation error to generate deterministic splits.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Gaussian mixture models for spots in microscopy using a new split/merge em algorithm

Pan

Kokaram

Hillebrand

et al. 2010

2010 IEEE International Conference on Image Processing

Self Cite

View full text Add to dashboard Cite

In confocal microscopy imaging, target objects are labeled with fluorescent markers in the living specimen, and usually appear as spots in the observed images. Spot detection and analysis is therefore an important task but it is still heavily reliant on manual analysis. In this paper, a novel shape modeling algorithm is proposed for automating the detection and analysis of the spots of interest. The algorithm exploits a Gaussian mixture model to characterize the spatial intensity distribution of the spots, and estimates parameters using a novel split-and-merge expectation maximization (SMEM) algorithm. In previous work the split step is random which is an issue for biological analysis where repeatability is important. The new split/merge steps are deterministic, hence more useful, and further do not impact adversely on the optimality of the final result.

show abstract

Immunochemical methods for identification of prey in freshwater zooplankton

Berges

Gronquist

Engevold

et al. 2020

Limnology & Ocean Methods

View full text Add to dashboard Cite

Large, predatory, non‐indigenous species such as the cladoceran, Bythotrephes longimanus, can have a profound effect on aquatic food webs, but determining the prey consumed by such species is problematic because they shred prey externally and consume only soft body tissues. We assessed the usefulness of an immunochemical approach, in which putative zooplankton prey were homogenized and polyclonal antibodies against total proteins were raised in rabbits and purified against predator proteins. Using either dot‐blot or enzyme‐linked immunosorbent assay‐plate formats, assays easily detected prey proteins in single B. longimanus. Issues encountered included some degree of cross‐reaction with predator proteins and among prey species for some antisera and questions concerning determination of a correct blank value for assays, however, there are feasible approaches to resolving these issues. Immunochemical assays were readily applied in to a total of 185 individual B. longimanus in the Lake Michigan pelagic ecosystem in summer of 2008 and 2009 and indicated ingestion of cladoceran and copepod species, typically described in the literature, but also dreissenid veliger larvae and the large copepod Limnocalanus macrurus, not usually considered prey for Bythotrephes. Assays results showed some degree of correlation with numerical abundances of zooplankton prey in the water column, suggesting generalist feeding. Protein immunochemical assays offer a quick, sensitive and practical approach to examining planktonic food‐web questions that are compatible with, but have certain advantages over, alternatives such as nucleic acid techniques.

show abstract

Gaussian mixtures for intensity modeling of spots in microscopy

Cited by 6 publications

References 10 publications

The Me31B DEAD-Box Helicase Localizes to Postsynaptic Foci and Regulates Expression of a CaMKII Reporter mRNA in Dendrites of Drosophila Olfactory Projection Neurons

The Me31B DEAD-Box Helicase Localizes to Postsynaptic Foci and Regulates Expression of a CaMKII Reporter mRNA in Dendrites of Drosophila Olfactory Projection Neurons

Gaussian mixture models for spots in microscopy using a new split/merge em algorithm

Immunochemical methods for identification of prey in freshwater zooplankton

Contact Info

Product

Resources

About