Dataset Information

High-throughput measurement of plant fitness traits with an object detection method using Faster R-CNN.

ABSTRACT: Revealing the contributions of genes to plant phenotype is frequently challenging because loss-of-function effects may be subtle or masked by varying degrees of genetic redundancy. Such effects can potentially be detected by measuring plant fitness, which reflects the cumulative effects of genetic changes over the lifetime of a plant. However, fitness is challenging to measure accurately, particularly in species with high fecundity and relatively small propagule sizes such as Arabidopsis thaliana. An image segmentation-based method using the software ImageJ and an object detection-based method using the Faster Region-based Convolutional Neural Network (R-CNN) algorithm were used for measuring two Arabidopsis fitness traits: seed and fruit counts. The segmentation-based method was error-prone (correlation between true and predicted seed counts, r² = 0.849) because seeds touching each other were undercounted. By contrast, the object detection-based algorithm yielded near perfect seed counts (r² = 0.9996) and highly accurate fruit counts (r² = 0.980). Comparing seed counts for wild-type and 12 mutant lines revealed fitness effects for three genes; fruit counts revealed the same effects for two genes. Our study provides analysis pipelines and models to facilitate the investigation of Arabidopsis fitness traits and demonstrates the importance of examining fitness traits when studying gene functions.

SUBMITTER: Wang P

PROVIDER: S-EPMC9310946 | biostudies-literature | 2022 May

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

High-throughput measurement of plant fitness traits with an object detection method using Faster R-CNN.

Wang Peipei P Meng Fanrui F Donaldson Paityn P Horan Sarah S Panchy Nicholas L NL Vischulis Elyse E Winship Eamon E Conner Jeffrey K JK Krysan Patrick J PJ Shiu Shin-Han SH Lehti-Shiu Melissa D MD

The New phytologist 20220326 4

Revealing the contributions of genes to plant phenotype is frequently challenging because loss-of-function effects may be subtle or masked by varying degrees of genetic redundancy. Such effects can potentially be detected by measuring plant fitness, which reflects the cumulative effects of genetic changes over the lifetime of a plant. However, fitness is challenging to measure accurately, particularly in species with high fecundity and relatively small propagule sizes such as Arabidopsis thalian ...[more]

PMID: 35218008

Similar Datasets

Project description:BackgroundAlthough diagnostic ultrasound can non-invasively capture the image of abdominal viscera, diagnosis of the continuous ultrasound liver images to detect a liver tumor effectively and to determine whether the detected is benign or malignant is nontrivial. In order to minimize the gaps in diagnostic accuracy depending on doctor's proficiency, we built an automated system to support the ultrasonography of liver tumors by employing deep learning technologies.MethodsWe constructed a neural network model for the automated detection of tumor tissues and blood vessels from the sequential liver ultrasound images. Faster region-based convolutional neural networks (Faster R-CNN) is employed as a base model for the object detection, which can output the detection results in 4 frames per second and enable the system to be particularly suitable for the real time ultrasonography. Moreover, we proposed a new neural network architecture feeding both the current and previous images into Faster R-CNN. For training the models, intraoperative ultrasound images obtained from one hepatocellular carcinoma (HCC) patient were used. The obtained image was a multifaceted observation of the liver and includes one HCC and some blood vessels. We labeled 91 images with the help of a liver specialist. We compared the tumor detection performance of the plain Faster R-CNN model with that of the proposed model.ResultsWe find that both the models performed well in detecting HCC and blood vessels, after training with 400 epochs using Adam. However, the mean precision of our model reaches 0.549, which is 0.019 better than that of the plain Faster R-CNN, and the mean sensitivity of our model about HCC reaches 0.623±0.385 for 30 scenes of sequential liver ultrasound images, which is also 0.146 better than that of the plain Faster R-CNN model.ConclusionsThe comparison between the proposed model and the plain Faster R-CNN model shows that we achieved better accuracy in tumor detection, in terms of the mean precision as well as the mean sensitivity, with the proposed model.

Project description:To evaluate changes in reproductive fitness of bacteria, e.g., after acquisition of antimicrobial resistance, a low-cost high-throughput method to analyze bacterial growth on agar is desirable for broad usability. In our bacterial quantitative fitness analysis (BaQFA), arrayed cultures are spotted on agar and photographed sequentially while growing. These time-lapse images are analyzed using a purpose-built open-source software to derive normalized image intensity (NI) values for each culture spot. Subsequently, a Gompertz growth model is fitted to NI values, and fitness is calculated from model parameters. To represent a range of clinically important pathogenic bacteria, we used different strains of Enterococcus faecium, Escherichia coli, and Staphylococcus aureus, with and without antimicrobial resistance. Relative competitive fitness (RCF) was defined as the mean fitness ratio of two strains growing competitively on one plate.BaQFA permitted the accurate construction of growth curves from bacteria grown on semisolid agar plates and fitting of Gompertz models. Normalized image intensity values showed a strong association with the total CFU/ml count per spotted culture (P < 0.001) for all strains of the three species. BaQFA showed relevant reproductive fitness differences between individual strains, suggesting substantially higher fitness of methicillin-resistant S. aureus JE2 than Cowan (RCF, 1.58; P < 0.001). Similarly, the vancomycin-resistant E. faecium ST172b showed higher competitive fitness than susceptible E. faecium ST172 (RCF, 1.59; P < 0.001). Our BaQFA method allows detection of fitness differences between bacterial strains and may help to estimate epidemiological antimicrobial persistence or contribute to the prediction of clinical outcomes in severe infections.IMPORTANCE Reproductive fitness of bacteria is a major factor in the evolution and persistence of antimicrobial resistance and may play an important role in severe infections. With a computational approach to quantify fitness in bacteria growing competitively on agar plates, our high-throughput method has been designed to obtain additional phenotypic data for antimicrobial resistance analysis at a low cost. Furthermore, our bacterial quantitative fitness analysis (BaQFA) enables the investigation of a link between bacterial fitness and clinical outcomes in severe invasive bacterial infections. This may allow future use of our method for patient management and risk stratification of clinical outcomes. Our proposed method uses open-source software and a hardware setup that can utilize consumer electronics. This will enable a wider community of researchers, including those from low-resource countries, where the burden of antimicrobial resistance is highest, to obtain valuable information about emerging bacterial strains.

Dataset Information

High-throughput measurement of plant fitness traits with an object detection method using Faster R-CNN.

Publications

High-throughput measurement of plant fitness traits with an object detection method using Faster R-CNN.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets