Receive a weekly summary and discussion of the top papers of the week by leading researchers in the field.

General General

PrismEXP: gene annotation prediction from stratified gene-gene co-expression matrices.

In PeerJ

BACKGROUND : Gene-gene co-expression correlations measured by mRNA-sequencing (RNA-seq) can be used to predict gene annotations based on the co-variance structure within these data. In our prior work, we showed that uniformly aligned RNA-seq co-expression data from thousands of diverse studies is highly predictive of both gene annotations and protein-protein interactions. However, the performance of the predictions varies depending on whether the gene annotations and interactions are cell type and tissue specific or agnostic. Tissue and cell type-specific gene-gene co-expression data can be useful for making more accurate predictions because many genes perform their functions in unique ways in different cellular contexts. However, identifying the optimal tissues and cell types to partition the global gene-gene co-expression matrix is challenging.

RESULTS : Here we introduce and validate an approach called PRediction of gene Insights from Stratified Mammalian gene co-EXPression (PrismEXP) for improved gene annotation predictions based on RNA-seq gene-gene co-expression data. Using uniformly aligned data from ARCHS4, we apply PrismEXP to predict a wide variety of gene annotations including pathway membership, Gene Ontology terms, as well as human and mouse phenotypes. Predictions made with PrismEXP outperform predictions made with the global cross-tissue co-expression correlation matrix approach on all tested domains, and training using one annotation domain can be used to predict annotations in other domains.

CONCLUSIONS : By demonstrating the utility of PrismEXP predictions in multiple use cases we show how PrismEXP can be used to enhance unsupervised machine learning methods to better understand the roles of understudied genes and proteins. To make PrismEXP accessible, it is provided via a user-friendly web interface, a Python package, and an Appyter. AVAILABILITY. The PrismEXP web-based application, with pre-computed PrismEXP predictions, is available from: https://maayanlab.cloud/prismexp; PrismEXP is also available as an Appyter: https://appyters.maayanlab.cloud/PrismEXP/; and as Python package: https://github.com/maayanlab/prismexp.

Lachmann Alexander, Rizzo Kaeli A, Bartal Alon, Jeon Minji, Clarke Daniel J B, Ma’ayan Avi

2023

Druggable genome, Gene expression, Gene function predictions, RNA-seq, Transcriptomics, Unsupervised learning

General General

WhoseEgg: classification software for invasive carp eggs.

In PeerJ

The collection of fish eggs is a commonly used technique for monitoring invasive carp. Genetic identification is the most trusted method for identifying fish eggs but is expensive and slow. Recent work suggests random forest models could provide an inexpensive method for identifying invasive carp eggs based on morphometric egg characteristics. While random forests provide accurate predictions, they do not produce a simple formula for obtaining new predictions. Instead, individuals must have knowledge of the R coding language, limiting the individuals who can use the random forests for resource management. We present WhoseEgg: a web-based point-and-click application that allows non-R users to access random forests via a point and click interface to rapidly identify fish eggs with an objective of detecting invasive carp (Bighead, Grass, and Silver Carp) in the Upper Mississippi River basin. This article provides an overview of WhoseEgg, an example application, and future research directions.

Goode Katherine, Weber Michael J, Dixon Philip M

2023

Bigheaded carp, Invasive species, Machine learning, Morphometrics, R Shiny, Random forests, Reproduction

General General

Integration of surface-enhanced Raman spectroscopy (SERS) and machine learning tools for coffee beverage classification.

In Digital Chemical Engineering

Surface-enhanced Raman spectroscopy (SERS) is a powerful tool for molecule identification. However, profiling complex samples remains a challenge because SERS peaks are likely to overlap, confounding features when multiple analytes are present in a single sample. In addition, SERS often suffers from high variability in signal enhancement due to nonuniform SERS substrate. The machine learning classification techniques widely used for facial recognition are excellent tools to overcome the complexity of SERS data interpretation. Herein, we reported a sensor for classifying coffee beverages by integrating SERS, feature extractions, and machine learning classifiers. A versatile and low-cost SERS substrate, called nanopaper, was used to enhance Raman signals of dilute compounds in coffee beverages. Two classic multivariate analysis techniques, Principal Component Analysis (PCA) and Discriminant Analysis of Principal Components (DAPC), were used to extract the significant spectral features, and the performance of various machine learning classifiers was evaluated. The combination of DAPC with Support Vector Machine (SVM) or K-Nearest Neighbor (KNN) shows the best performance for classifying coffee beverages. This user-friendly and versatile sensor has the potential to be a practical quality-control tool for the food industry.

Hu Qiang, Sellers Chase, Kwon Joseph Sang-Il, Wu Hung-Jen

2022-Jun

Classification, Coffee, Feature extraction, Machine learning, Surface-enhanced Raman spectroscopy (SERS)

Ophthalmology Ophthalmology

Morphological characteristics of retinal vessels in eyes with high myopia: Ultra-wide field images analyzed by artificial intelligence using a transfer learning system.

In Frontiers in medicine

PURPOSE : The purpose of this study is to investigate the retinal vascular morphological characteristics in high myopia patients of different severity.

METHODS : 317 eyes of high myopia patients and 104 eyes of healthy control subjects were included in this study. The severity of high myopia patients is classified into C0-C4 according to the Meta Analysis of the Pathologic Myopia (META-PM) classification and their vascular morphological characteristics in ultra-wide field imaging were analyzed using transfer learning methods and RU-net. Correlation with axial length (AL), best corrected visual acuity (BCVA) and age was analyzed. In addition, the vascular morphological characteristics of myopic choroidal neovascularization (mCNV) patients and their matched high myopia patients were compared.

RESULTS : The RU-net and transfer learning system of blood vessel segmentation had an accuracy of 98.24%, a sensitivity of 71.42%, a specificity of 99.37%, a precision of 73.68% and a F1 score of 72.29. Compared with healthy control group, high myopia group had smaller vessel angle (31.12 ± 2.27 vs. 32.33 ± 2.14), smaller fractal dimension (Df) (1.383 ± 0.060 vs. 1.424 ± 0.038), smaller vessel density (2.57 ± 0.96 vs. 3.92 ± 0.93) and fewer vascular branches (201.87 ± 75.92 vs. 271.31 ± 67.37), all P < 0.001. With the increase of myopia maculopathy severity, vessel angle, Df, vessel density and vascular branches significantly decreased (all P < 0.001). There were significant correlations of these characteristics with AL, BCVA and age. Patients with mCNV tended to have larger vessel density (P < 0.001) and more vascular branches (P = 0.045).

CONCLUSION : The RU-net and transfer learning technology used in this study has an accuracy of 98.24%, thus has good performance in quantitative analysis of vascular morphological characteristics in Ultra-wide field images. Along with the increase of myopic maculopathy severity and the elongation of eyeball, vessel angle, Df, vessel density and vascular branches decreased. Myopic CNV patients have larger vessel density and more vascular branches.

Mao Jianbo, Deng Xinyi, Ye Yu, Liu Hui, Fang Yuyan, Zhang Zhengxi, Chen Nuo, Sun Mingzhai, Shen Lijun

2022

choroidal neovascularization, deep learning, high myopia, ultra-wide field imaging, vascular morphology

General General

Impact evaluation and economic benefit analysis of a domestic violence and abuse UK police intervention.

In Frontiers in psychology ; h5-index 92.0

This study evaluated the impact and economic benefit of Cautioning and Relationship Abuse (CARA), an intervention which aims to reduce re-offending of first-time low-level domestic violence and abuse perpetrators. The analysis was based on two samples drawn from separate UK police force areas. CARA's impact was assessed using a matched sample of similar offenders from a time when CARA was not available. The matching was based on a host of offender and victim characteristics and machine learning methods were employed. The results show that the CARA intervention has a significant impact on the amount of recidivism but no significant reduction in the severity of the crimes. The benefit-cost ratio in both police force areas is greater than one and estimated to be 2.75 and 11.1, respectively, across the two police force areas. Thus, for each pound (£) invested in CARA, there is an economic benefit of 2.75-11.1 pounds, annually.

Karavias Yiannis, Bandyopadhyay Siddhartha, Christie Christine, Bradbury-Jones Caroline, Taylor Julie, Kane Eddie, Flowe Heather D

2023

Crime Harm Index, batterer intervention program, domestic violence and abuse, economic evaluation, evidence-based policing, intimate partner violence, machine learning

General General

Deep Learning Models for Cystoscopic Recognition of Hunner Lesion in Interstitial Cystitis.

In European urology open science

BACKGROUND : Accurate cystoscopic recognition of Hunner lesions (HLs) is indispensable for better treatment prognosis in managing patients with Hunner-type interstitial cystitis (HIC), but frequently challenging due to its varying appearance.

OBJECTIVE : To develop a deep learning (DL) system for cystoscopic recognition of a HL using artificial intelligence (AI).

DESIGN SETTING AND PARTICIPANTS : A total of 626 cystoscopic images collected from January 8, 2019 to December 24, 2020, consisting of 360 images of HLs from 41 patients with HIC and 266 images of flat reddish mucosal lesions resembling HLs from 41 control patients including those with bladder cancer and other chronic cystitis, were used to create a dataset with an 8:2 ratio of training images and test images for transfer learning and external validation, respectively. AI-based five DL models were constructed, using a pretrained convolutional neural network model that was retrained to output 1 for a HL and 0 for control. A five-fold cross-validation method was applied for internal validation.

OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS : True- and false-positive rates were plotted as a receiver operating curve when the threshold changed from 0 to 1. Accuracy, sensitivity, and specificity were evaluated at a threshold of 0.5. Diagnostic performance of the models was compared with that of urologists as a reader study.

RESULTS AND LIMITATIONS : The mean area under the curve of the models reached 0.919, with mean sensitivity of 81.9% and specificity of 85.2% in the test dataset. In the reader study, the mean accuracy, sensitivity, and specificity were, respectively, 83.0%, 80.4%, and 85.6% for the models, and 62.4%, 79.6%, and 45.2% for expert urologists. Limitations include the diagnostic nature of a HL as warranted assertibility.

CONCLUSIONS : We constructed the first DL system that recognizes HLs with accuracy exceeding that of humans. This AI-driven system assists physicians with proper cystoscopic recognition of a HL.

PATIENT SUMMARY : In this diagnostic study, we developed a deep learning system for cystoscopic recognition of Hunner lesions in patients with interstitial cystitis. The mean area under the curve of the constructed system reached 0.919 with mean sensitivity of 81.9% and specificity of 85.2%, demonstrating diagnostic accuracy exceeding that of human expert urologists in detecting Hunner lesions. This deep learning system assists physicians with proper diagnosis of a Hunner lesion.

Iwaki Takuya, Akiyama Yoshiyuki, Nosato Hirokazu, Kinjo Manami, Niimi Aya, Taguchi Satoru, Yamada Yuta, Sato Yusuke, Kawai Taketo, Yamada Daisuke, Sakanashi Hidenori, Kume Haruki, Homma Yukio, Fukuhara Hiroshi

2023-Mar

Artificial intelligence, Bladder pain syndrome, Deep learning, Hunner lesion, Interstitial cystitis