A multi-omics machine learning framework in predicting the survival of colorectal cancer patients

Colorectal cancer (CRC) is one of the most common cancers worldwide. Accurate early detection and diagnosis, comprehensive assessment of treatment response, and precise prediction of prognosis are essential to improve the patients’ survival rate. In recent years, due to the explosion of clinical and omics data, and groundbreaking research in machine learning, artificial intelligence (AI) has shown a great application potential in clinical field of CRC, providing new auxiliary approaches for clinicians to identify high-risk patients, select precise and personalized treatment plans, as well as to predict prognoses. This review comprehensively analyzes and summarizes the research progress and clinical application value of AI technologies in CRC screening, diagnosis, treatment, and prognosis, demonstrating the current status of the AI in the main clinical stages. The limitations, challenges, and future perspectives in the clinical implementation of AI are also discussed.

https://www.mdpi.com/1718-7729/29/3/146/pdf?version=1646664545

Applications of Artificial Intelligence in Screening, Diagnosis, Treatment, and Prognosis of Colorectal Cancer

REGLIV: Molecular regulation data of diverse living systems facilitating current multiomics research

MOTIVATION
Tumor mutational burden (TMB) is an indicator of the efficacy and prognosis of immune checkpoint therapy in colorectal cancer (CRC). Cancer patients with high TMB (TMB_H) values tend to benefit from immunotherapy, whereas those with low TMB (TMB_L) values tend to be not. Though whole-exome sequencing (WES) is considered the gold standard for determining TMB, it is difficult to be applied in clinical practice due to its high cost. There are also a few DNA panel-based methods to estimate TMB; however, their detection cost is also high, and the associated wet-lab experiments usually take days, which emphasize the need for faster and cheaper alternatives.


METHODS
In this study, we propose a multi-modal deep learning model based on a residual network (ResNet) and multi-modal compact bilinear pooling to predict TMB status (i.e., TMB_H or TMB_L) directly from histopathological images and clinical data. We applied the model to CRC data from The Cancer Genome Atlas and compared it with four other popular methods, namely, ResNet18, ResNet50, VGG19, and AlexNet. We tested different TMB thresholds, namely, percentiles of 10%, 14.3%, 15%, 16.3%, 20%, 30% and 50%, to differentiate TMB_H and TMB_L.


RESULTS
For the percentile 14.3% (i.e., TMB value 20) and ResNet18, our model achieved an area under the receiver operating characteristic curve of 0.817 after five-fold cross-validation, which was better than that of other compared models. In addition, we also found that TMB values were significantly associated with tumor stage and N and M stages. Our study shows that deep learning models can predict TMB status from histopathological images and clinical information only, which is worth clinical application.

/pdf/predicting-colorectal-cancer-tumor-mutational-burden-from-310qoup3.pdf

Predicting colorectal cancer tumor mutational burden from histopathological images and clinical information using multi-modal deep learning

Plankton biodiversity serves as ecological indicator of aquatic environment due to their response to environmental changes. Study was carried out for of 6 months (May – October) in Dadin Kowa dam, Gombe state Nigeria. Samples collected using plankton net into a bottles preserve in 5% formalin then transported to the laboratory for identification, counting and for further analysis. SPSS (v.11.0.1) was used for data analysis.  Phytoplanktons group recorded Chlorophyta highest (1,960) and lowest (860) in May and July. Bacillariophyta maximum (990) and minimum (450) number in June and September, Cyanophyta maximum (450) in July and Minimum (20) in September. Euglenophyta was only recorded in September (30), Cryptophyta observed only in June (20) and September (20), Dinophyta highest in May and September (40) minimum (10) in August. Xantophyta minimum (10) in August and maximum (30) in October. The Zooplanktons include Brightwellii highest (428) in June lowest (121) in October, Monia highest (97) and lowest (30) in June and September, Diaptomuus highest (112) in May and the lowest (30) in September. Philodina highest (550) in June while lowest (195) in September, Daphnia highest (43) and lowest (26) in May and September. Cyclops highest (70) in July while lowest (35) in September and October, Bdelloid highest (402) in May and lowest (116) in October. Camtocerus highest (76) in May and lowest (35) in September and October, Bidentata highest (331) in May and lowest (49) in October. Torrential rains,

/pdf/plankton-abundance-and-diversity-in-dadin-kowa-dam-in-gombe-p0vkyk3cx3.pdf

Plankton abundance and diversity in dadin kowa dam in gombe state, nigeria

Machine learning and network-based models to identify genetic risk factors to the progression and survival of colorectal cancer.

The present study was carried out from May, 2011 to March, 2012 which deals with the abundance of zooplankton in Ramsagardighi, Dinajpur, Bangladesh. A total of 27 species of 21 genera were recorded in which rotifers were more abundant with 9 genera and 12species followed by copepods with 6 genera and 7 species and cladocerans with 5 genera and 7 species from the study area. A crustacean, nauplius larva was also recorded from the study area. Monthly fluctuation of some physico-chemical parameters was also noted. Bangladesh J. Zool. 43 (2): 303-312, 2015

/pdf/abundance-of-zooplankton-in-ramsagar-dighi-dinajpur-3nz05xn1su.pdf

Jakir Hossain

Papers

Machine learning and network-based models to identify genetic risk factors to the progression and survival of colorectal cancer.

Abundance of Zooplankton in Ramsagar-Dighi, Dinajpur, Bangladesh