Machine learning techniques in disease forecasting: a case study on rice blast prediction

Kaundal, Rakesh; Kapoor, Amar S; Raghava, Gajendra P. S.

doi:10.1186/1471-2105-7-485

Cited by 137 publications

(43 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…SVMs are a class of supervised learning algorithms based on the optimization principle from statistical learning theory [ 28 , 29 ]. Support vector machines have been used widely in computational biology in diverse topics such as subcellular localization [ 18 , 30 - 32 ], protein function prediction [ 33 ], secondary structure prediction [ 34 ], disease forecasting [ 35 ]. SVMs solve classification problems by calculating a hyperplane that separates the training data with a maximum margin.…”

Section: Methodsmentioning

confidence: 99%

LacSubPred: predicting subtypes of Laccases, an important lignin metabolism-related enzyme class, using in silico approaches

et al. 2014

Self Cite

View full text Add to dashboard Cite

BackgroundLaccases (E.C. 1.10.3.2) are multi-copper oxidases that have gained importance in many industries such as biofuels, pulp production, textile dye bleaching, bioremediation, and food production. Their usefulness stems from the ability to act on a diverse range of phenolic compounds such as o-/p-quinols, aminophenols, polyphenols, polyamines, aryl diamines, and aromatic thiols. Despite acting on a wide range of compounds as a family, individual Laccases often exhibit distinctive and varied substrate ranges. This is likely due to Laccases involvement in many metabolic roles across diverse taxa. Classification systems for multi-copper oxidases have been developed using multiple sequence alignments, however, these systems seem to largely follow species taxonomy rather than substrate ranges, enzyme properties, or specific function. It has been suggested that the roles and substrates of various Laccases are related to their optimal pH. This is consistent with the observation that fungal Laccases usually prefer acidic conditions, whereas plant and bacterial Laccases prefer basic conditions. Based on these observations, we hypothesize that a descriptor-based unsupervised learning system could generate homology independent classification system for better describing the functional properties of Laccases.ResultsIn this study, we first utilized unsupervised learning approach to develop a novel homology independent Laccase classification system. From the descriptors considered, physicochemical properties showed the best performance. Physicochemical properties divided the Laccases into twelve subtypes. Analysis of the clusters using a t-test revealed that the majority of the physicochemical descriptors had statistically significant differences between the classes. Feature selection identified the most important features as negatively charges residues, the peptide isoelectric point, and acidic or amidic residues. Secondly, to allow for classification of new Laccases, a supervised learning system was developed from the clusters. The models showed high performance with an overall accuracy of 99.03%, error of 0.49%, MCC of 0.9367, precision of 94.20%, sensitivity of 94.20%, and specificity of 99.47% in a 5-fold cross-validation test. In an independent test, our models still provide a high accuracy of 97.98%, error rate of 1.02%, MCC of 0.8678, precision of 87.88%, sensitivity of 87.88% and specificity of 98.90%.ConclusionThis study provides a useful classification system for better understanding of Laccases from their physicochemical properties perspective. We also developed a publically available web tool for the characterization of Laccase protein sequences (http://lacsubpred.bioinfo.ucr.edu/). Finally, the programs used in the study are made available for researchers interested in applying the system to other enzyme classes (https://github.com/tweirick/SubClPred).

show abstract

Section: Methodsmentioning

confidence: 99%

LacSubPred: predicting subtypes of Laccases, an important lignin metabolism-related enzyme class, using in silico approaches

et al. 2014

Self Cite

View full text Add to dashboard Cite

show abstract

“…oryzae strains at the maximum tillering stage (Mew et al 1993). Rice blast, caused by the fungus Magnaporthe oryzae , is the most important fungal disease in rice production and its repercussion is the yield loss of 157 million tons of rice annually in the world (Kaundal et al 2006). During the evolution, rice has co-evolved disease resistance ( R ) genes against the infection by the two kinds of pathogens (Liu et al 2014).…”

Section: Introductionmentioning

confidence: 99%

Marker-assisted breeding of the rice restorer line Wanhui 6725 for disease resistance, submergence tolerance and aromatic fragrance

Luo

Zhang

et al. 2016

Rice

View full text Add to dashboard Cite

BackgroundRice is a staple food crop in the world. With the increase in world population and economic development, farmers need to produce more rice in limited field. However, the rice production is frequently affected by biotic and abiotic stresses. The use of natural disease resistance and stress tolerance through genetic breeding is the most efficient and economical way to combat or acclimate to these stresses. In addition, rice with aromatic fragrance can significantly increase market value for its good grain quality. Mianhui 725 (MH725) is an elite restorer line that has been widely used to produce three-line hybrid rice in China. We previously introduced rice bacterial blight resistance genes Xa4 and Xa21 into MH725 and obtained an introgression rice line Wanhui 421 (WH421), which theoretically possesses 96.9% genetic background of MH725.ResultsHere we report the introduction and pyramiding of disease resistance genes Xa27 and Pi9, submergence tolerance gene Sub1A and aromatic fragrance gene badh2.1 in WH421 through backcrossing and marker-assisted selection. The newly developed introgression rice line was designated as Wanhui 6725 (WH6725), which theoretically possesses 95.0% genetic background of MH725. WH6725 and its hybrid rice conferred disease resistance to both blast and bacterial blight diseases and showed tolerance to submergence for over 14 days without significant loss of viability. Compared with non-aromatic rice MH725, WH6725 has strong aromatic fragrance. The major important agronomic traits and grain quality of WH6725 and its hybrid rice obtained in field trials were similar to those of MH725 and the control hybrid rice, indicating that WH6725 is as good as MH725 when it is used as a restorer line for three-line hybrid rice production.ConclusionWe have successfully developed a new restorer line WH6725 with disease resistance to rice blast and bacterial blight, tolerance to submergence and aromatic fragrance, which can be used to replace MH725 for hybrid rice production.

show abstract

“…In recent years, diverse modelling approaches such as control strategies for the prediction of fungal diseases have been proposed. Among them, linear regression models, the autoregressive integrated model of running mean time-series, and neural network models were applied in potato, grapevine, rice, and wheat [8,35,[37][38][39]. As a first approximation, it can be concluded that the combination of aerobiological data with weather data (specially wet periods) collected during nine crop cycles in A Limia was efficient in that it could predict several days of attack in advance during the development of the crop.…”

Section: Discussionmentioning

confidence: 99%

Interrupted Wet Period (IWP) to Forecast the Aerial Alternaria in Potato Crops of A Limia (Spain)

et al. 2019

View full text Add to dashboard Cite

Potato early blight caused by Alternaria solani generates significant economic losses in crops worldwide. Forecasting the risk of infection on crops is indispensable for the management of the fungal disease, ensuring maximum economic benefit but with minimal environmental impact. This work aimed to calculate the interrupted wet periods (IWP) according to the climate conditions of A Limia (Northwest of Spain) to optimize the prediction against early blight in potatoes. The study was performed during nine crop cycles. The relative hourly humidity and Alternaria concentration in the crop environment were taken into account. Alternaria levels were monitored by aerobiological techniques using a LANZONI VPPS-2000 volumetric trap. The relationships between weather conditions and airborne Alternaria concentration were statistically analyzed using Spearman correlations. To establish the effectiveness of wetness periods, the first important Alternaria peak was taken into account in each crop cycle (with a concentration greater than 70 spores/m 3 ). Considering the six interrupted wet periods of the system, it was possible to predict the first peak of Alternaria several days in advance (between 6 and 38 days), except in 2007 and 2018. Automated systems to predict the initiation of early blight in potato crop, such as interrupted wet periods, could be an effective basis for developing decision support systems. The incorporation of aerobiological data for the calculation of interrupted wet periods improved the results of this system.

show abstract

Machine learning techniques in disease forecasting: a case study on rice blast prediction

Cited by 137 publications

References 32 publications

LacSubPred: predicting subtypes of Laccases, an important lignin metabolism-related enzyme class, using in silico approaches

LacSubPred: predicting subtypes of Laccases, an important lignin metabolism-related enzyme class, using in silico approaches

Marker-assisted breeding of the rice restorer line Wanhui 6725 for disease resistance, submergence tolerance and aromatic fragrance

Interrupted Wet Period (IWP) to Forecast the Aerial Alternaria in Potato Crops of A Limia (Spain)

Contact Info

Product

Resources

About