Dataset Information

Versatile method for quantifying and analyzing morphological differences in experimentally obtained images.

ABSTRACT: Analyzing high-resolution images to gain insight into anatomical properties is an essential tool for investigation in many scientific fields. In plant biology, studying plant phenotypes from micrographs is often used to build hypotheses on gene function. In this report, we discuss a bespoke method for inspecting the significance in the differences between the morphologies of several plant mutants at cellular level. By examining a specific example in the literature, we will detail the approach previously used to quantify the effects of two gene families on the vascular development of hypocotyls in Arabidopsis thaliana. The method incorporates a MATLAB algorithm and statistical tools which can be modified and enhanced to tailor to different research questions in future studies.

SUBMITTER: Bagdassarian KS

PROVIDER: S-EPMC7012139 | biostudies-literature | 2020

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Versatile method for quantifying and analyzing morphological differences in experimentally obtained images.

Bagdassarian Kristine S KS Connor Katherine A KA Jermyn Ian H IH Etchells J Peter JP

Plant signaling & behavior 20191124 1

Analyzing high-resolution images to gain insight into anatomical properties is an essential tool for investigation in many scientific fields. In plant biology, studying plant phenotypes from micrographs is often used to build hypotheses on gene function. In this report, we discuss a bespoke method for inspecting the significance in the differences between the morphologies of several plant mutants at cellular level. By examining a specific example in the literature, we will detail the approach pr ...[more]

PMID: 31762388

Similar Datasets

Project description:BackgroundAside from single nucleotide polymorphisms, copy number variations (CNVs) are the most important factors in susceptibility to genetic disorders because they affect expression levels of genes. In previous studies, pyrosequencing, mini-sequencing, real-time PCR, invader assays and other techniques have been used to detect CNVs. However, the higher the copy number in a genome, the more difficult it is to resolve the copies, so a more accurate method for measuring CNVs and assigning genotype is needed.ResultsPCR followed by a quantitative oligonucleotide ligation assay (qOLA) was developed for quantifying CNVs. The accuracy and precision of the assay were evaluated for porcine KIT, which was selected as a model locus. Overall, the root mean squares of bias and standard deviation of qOLA were 2.09 and 0.45, respectively. These values are less than half of those in the published pyrosequencing assay for analyzing CNV in porcine KIT. Using a combined method of qOLA and another pyrosequencing for quantitative analysis of KIT copies with spliced forms, we confirmed the segregation of KIT alleles in 145 F1 animals with pedigree information and verified the correct assignment of genotypes. In a diagnostic test on 100 randomly sampled commercial pigs, there was perfect agreement between the genotypes obtained by grouping observations on a scatter plot and by clustering using the nearest centroid sorting method implemented in PROC FASTCLUS of the SAS package. In a test on 159 Large White pigs, there were only two discrepancies between genotypes assigned by the two clustering methods (98.7% agreement), confirming that the quantitative ligation assay established here makes genotyping possible through the accurate measurement of high KIT copy numbers (>4 per diploid genome). Moreover, the assay is sensitive enough for use on DNA from hair follicles, indicating that DNA from various sources could be used.ConclusionWe have established a high resolution quantification method using an oligonucleotide ligation assay to measure CNVs, and verified the reliability of genotype assignment for random animal samples using the nearest centroid sorting method. This new method will make it more practical to determine KIT CNV and to genotype the complicated Dominant White/KIT locus in pigs. This procedure could have wide applications for studying gene or segment CNVs in other species.

Project description:PurposeTo determine whether convolutional neural networks can detect morphological differences between images of microbiologically positive and negative corneal ulcers.MethodsA cross-sectional comparison of prospectively collected data consisting of bacterial and fungal cultures and smears from eyes with acute infectious keratitis at Aravind Eye Hospital. Two convolutional neural network architectures (DenseNet and MobileNet) were trained using images obtained from handheld cameras collected from culture-positive and negative images and smear-positive and -negative images. Each architecture was trained on two image sets: (1) one with labels assigned using only culture results and (2) one using culture and smear results. The outcome measure was area under the receiver operating characteristic curve for predicting whether an ulcer would be microbiologically positive or negative.ResultsThere were 1970 images from 886 patients were included. None of the models were better than random chance at predicting positive microbiologic results (area under the receiver operating characteristic curve ranged from 0.49 to 0.56; all confidence intervals included 0.5).ConclusionsThese two state-of-the-art deep convolutional neural network architectures could not reliably predict whether a corneal ulcer would be microbiologically positive or negative based on clinical photographs. This absence of detectable morphological differences informs the future development of computer vision models trained to predict the causative agent in infectious keratitis using corneal photography.Translational relevanceThese deep learning models were not able to identify morphological differences between microbiologically positive and negative corneal ulcers. This finding suggests that similar artificial intelligence models trained to identify the causative pathogen using only microbiologically positive cases may have potential to generalize well, including to cases with falsely negative microbiologic testing.

Dataset Information

Versatile method for quantifying and analyzing morphological differences in experimentally obtained images.

Publications

Versatile method for quantifying and analyzing morphological differences in experimentally obtained images.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets