Missingness Adapted Group Informed Clustered (MAGIC)-LASSO: A novel paradigm for prediction in data with widespread non-random missingness

The availability of large-scale biobanks linking rich phenotypes and biological measures are a powerful opportunity for scientific discovery. However, real-world collections frequently have extensive non-random missing data. Machine learning methods are able to predict missing data but performance is significantly impaired by block-wise missingness inherent to many biobanks. To address this, we developed Missingness Adapted Group-wise Informed Clustered LASSO (MAGIC-LASSO) which performs hierarchical clustering of variables based on missingness followed by sequential Group LASSO within clusters. Variables are pre-filtered for missingness and balance between training and target sets with final models built using stepwise inclusion of features ranked by completeness. This research has been conducted using the UK Biobank (n>500k) to predict unmeasured Alcohol Use Disorders Identification Test (AUDIT.) The phenotypic correlation between measured and predicted total score was 0.67 while genetic correlations between independent subjects was >0.86, demonstrating the method has significant accuracy and utility.

Features Clustering Around Latent Variables for High Dimensional Data

Clustering of variables is the task of grouping similar variables into different groups. It may be useful in several situations such as dimensionality reduction, feature selection, and detect redundancies. In the present study, we combine two methods of features clustering the clustering of variables around latent variables (CLV) algorithm and the k-means based co-clustering algorithm (kCC). Indeed, classical CLV cannot be applied to high dimensional data because this approach becomes tedious when the number of features increases.

New Approach Studying Interactions Regarding Trade-Off between Beef Performances and Meat Qualities

The beef cattle industry is facing multiple problems, from the unequal distribution of added value to the poor matching of its product with fast-changing demand. Therefore, the aim of this study was to examine the interactions between the main variables, evaluating the nutritional and organoleptic properties of meat and cattle performances, including carcass properties, to assess a new method of managing the trade-off between these four performance goals. For this purpose, each variable evaluating the parameters of interest has been statistically modeled and based on data collected on 30 Blonde d’Aquitaine heifers. The variables were obtained after a statistical pre-treatment (clustering of variables) to reduce the redundancy of the 62 initial variables. The sensitivity analysis evaluated the importance of each independent variable in the models, and a graphical approach completed the analysis of the relationships between the variables. Then, the models were used to generate virtual animals and study the relationships between the nutritional and organoleptic quality. No apparent link between the nutritional and organoleptic properties of meat (r = −0.17) was established, indicating that no important trade-off between these two qualities was needed. The 30 best and worst profiles were selected based on nutritional and organoleptic expectations set by a group of experts from the INRA (French National Institute for Agricultural Research) and Institut de l’Elevage (French Livestock Institute). The comparison between the two extreme profiles showed that heavier and fatter carcasses led to low nutritional and organoleptic quality.