New algorithms for detecting multi-effect and multi-way epistatic interactions

Abstract

Motivation Epistasis, which is the phenomenon of genetic interactions, plays a central role in many scientific discoveries. However, due to the combinatorial nature of the problem, it is extremely challenging to decipher the exact combinations of genes that trigger the epistatic effects. Many existing methods only focus on two-way interactions. Some of the most effective methods used machine learning techniques, but many were designed for special case-and-control studies or suffer from overfitting. We propose three new algorithms for multi-effect and multi-way epistases detection, with one guaranteeing global optimality and the other two being local optimization oriented heuristics. Results The computational performance of the proposed heuristic algorithm was compared with several state-of-the-art methods using a yeast dataset. Results suggested that searching for the global optimal solution could be extremely time consuming, but the proposed heuristic algorithm was much more effective and efficient than others at finding a close-to-optimal solution. Moreover, it was able to provide biological insight on the exact configurations of epistases, besides achieving a higher prediction accuracy than the state-of-the-art methods. Availability and implementation Data source was publicly available and details are provided in the text.

Maize yield and nitrate loss prediction with machine learning algorithms

“…Traditionally, farmers rely on their experiences and past historical data such as the crop yields and weather to make important decisions to increase short-term profitability and long-term sustainability of their operation (Arbuckle and Rosman 2014). New promising technologies such as machine learning (ML) have emerged over the last years that can potentially aid farmers' decision making (Hoogenboom et al 2004, González Sánchez et al 2014, Togliatti et al 2017, Basso and Liu 2018, Ansarifar and Wang 2018, Moeinizade et al 2019. However, the lack of spatial and temporal data that cover a range of production (e.g.…”

Performance prediction of crosses in plant breeding through genotype by environment interactions

“…Three hyperparameters were tuned using fivefold cross validation (without data leakage): “nrounds”, “eta”, and “gamma”. A G E interactions model 33 was implemented in MATLAB (Version 2018a), which used heuristic algorithms to detect multi-way and multi-effect epistasis (interactions between binary variables). It is equivalent to the G E interactions model without integrating with the random forest models.…”

“…The G E interactions model was designed to detect interactions among specific hybrid, location, and weather variables. This model is built off of a recently published algorithm 33 , which was designed to detect genetic interactions in the form of epistases. The algorithm was found to be effective in detecting multiple interactions involving multiple variables.…”

“…In our model, a similar approach is used to detect interactions among hybrid, soil, and weather at a much larger scale with and . To overcome the computational challenges, we used a similar heuristic algorithm as in 33 , which had three desirable features: (1) it used cross-validation to avoid-overfitting; (2) it was able to find local optimal solutions efficiently; and (3) it could be parameterized to balance computation time and solution quality.…”

“…The random forest model 32 is known for its capability to approximate general form nonlinear relationships among the variables. On the other hand, the interaction-detection model originated from a recently published algorithm 33 that has been shown to be particularly effective in detecting epistatic type of interactions. Our model extends that algorithm to the detection of genotype by environment interactions (G E).…”