Bayesian prediction of bacterial growth temperature range based on genome sequences

Jensen, Dan Børge; Vesth, Tammi Camilla; Hallin, Peter F.; Pedersen, Anders Gorm; Ussery, David W.

doi:10.1186/1471-2164-13-s7-s3

Cited by 26 publications

(24 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This approach resulted in a training dataset with 5,532 organisms annotated with OGT, as well as a dataset with 1,438 un-annotated organisms. The annotated training dataset comprises 4,974 bacteria, 222 archaea and 337 eukarya ( Figure 1) and is much larger than those used in other approaches, such as 22 bacteria(30 ), 77 bacteria (33 ) or 204 prokaryotes (32 ). In the annotated dataset the number of proteins in each organism follows a normal distribution centered around 3,000 (Figure 1i).…”

Section: Resultsmentioning

confidence: 99%

“…Additionally, Zeldovich found that the sum fraction of the seven amino acids I, V, Y, W, R, E and L showed a correlation coefficient as high as 0.93 with OGT in a dataset consisting of 204 proteomes of archaea and bacteria (32 ). Jensen et al developed a Bayesian classifier to distinguish three thermophilicity classes (thermophiles, mesophiles and psychrophiles) based on 77 bacteria with known OGT (33 ). Training datasets containing the OGTs for a large number of organisms have been hard to obtain, something which has prevented the development of state-of-the-art machine learning models for OGT prediction.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Machine learning applied to predicting microorganism growth temperatures and enzyme catalytic optima

Rabe

Nielsen

et al. 2019

Preprint

View full text Add to dashboard Cite

Enzymes that catalyze chemical reactions at high temperatures are used for industrial biocatalysis, applications in molecular biology, and as highly evolvable starting points for protein engineering. The optimal growth temperature (OGT) of organisms is commonly used to estimate the stability of enzymes encoded in their genomes, but the number of experimentally determined OGT values are limited, particularly for thermophilic organisms. Here, we report on the development of a machine learning model that can accurately predict OGT for bacteria, archaea and microbial eukaryotes directly from their proteome-wide 2-mer amino acid composition. The trained model is made freely available for re-use. In a subsequent step we OGT data in combination with amino acid composition of individual enzymes to develop a second machine learning model -for prediction of enzyme catalytic temperature optima (T opt ). The resulting model generates enzyme T opt estimates that are far superior to using OGT alone. Finally, we predict T opt for 6.5 million enzymes, covering 4,447 enzyme classes, and make the resulting dataset available for researchers. This work enables simple and rapid identification of enzymes that are potentially functional at extreme temperatures.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Machine learning applied to predicting microorganism growth temperatures and enzyme catalytic optima

Rabe

Nielsen

et al. 2019

Preprint

View full text Add to dashboard Cite

show abstract

“…In general, cellular processes speed up as temperature increases, but extremely high temperatures can also denature proteins and negatively affect biochemical reactions. Proteins function best within a specific temperature window that maximizes enzymatic reaction rate without denaturing the protein, and most enzymes have evolved within an optimal temperature range that is closely tied to environmental temperature (1) . Few enzymes show optimal activity more than 10°C above or below the optimal growth temperature of the host organism (1,2) .…”

Section: Introductionmentioning

confidence: 99%

Building a tRNA thermometer to access the world’s biochemical diversity

Çimen

Buckler

2020

Preprint

View full text Add to dashboard Cite

ABSTRACTBecause ambient temperature affects biochemical reactions, organisms living in extreme temperature conditions adapt protein composition and structure to maintain biochemical functions. While it is not feasible to experimentally determine optimal growth temperature (OGT) for every known microbial species, organisms adapted to different temperatures have measurable differences in DNA, RNA, and protein composition that allow OGT prediction from genome sequence alone. In this study, we built a model using tRNA sequence to predict OGT. We used tRNA sequences from 100 archaea and 683 bacteria species as input to train two Convolutional Neural Network models. The first pairs individual tRNA sequences from different species to predict which comes from a more thermophilic organism, with accuracy ranging from 0.538 to 0.992. The second uses the complete set of tRNAs in a species to predict optimal growth temperature, achieving a maximum r2 of 0.86; comparable with other prediction accuracies in the literature despite a significant reduction in the quantity of input data. This model improves on previous OGT prediction models by providing a model with minimum input data requirements, removing laborious feature extraction and data preprocessing steps, and widening the scope of valid downstream analyses.

show abstract

“…The NBC is a relatively simple classification method, but it has been shown to be useful in a wide range of fields, such as prediction of bacterial thermophilicity (Jensen et al, 2012), diagnosis of classical swine fewer (Geenen et al, 2011), and detection of clinical mastitis (Steeneveld et al, 2009). The NBC has advantages over comparable classification methods, such as artificial neural networks or logistic regression functions, because missing observations can be easily handled in an NBC by including only the observations that are available.…”

Section: Introductionmentioning

confidence: 99%

Bayesian integration of sensor information and a multivariate dynamic linear model for prediction of dairy cow mastitis

Jensen

Hogeveen

Vries

2016

Journal of Dairy Science

View full text Add to dashboard Cite

Rapid detection of dairy cow mastitis is important so corrective action can be taken as soon as possible. Automatically collected sensor data used to monitor the performance and the health state of the cow could be useful for rapid detection of mastitis while reducing the labor needs for monitoring. The state of the art in combining sensor data to predict clinical mastitis still does not perform well enough to be applied in practice. Our objective was to combine a multivariate dynamic linear model (DLM) with a naïve Bayesian classifier (NBC) in a novel method using sensor and nonsensor data to detect clinical cases of mastitis. We also evaluated reductions in the number of sensors for detecting mastitis. With the DLM, we co-modeled 7 sources of sensor data (milk yield, fat, protein, lactose, conductivity, blood, body weight) collected at each milking for individual cows to produce one-step-ahead forecasts for each sensor. The observations were subsequently categorized according to the errors of the forecasted values and the estimated forecast variance. The categorized sensor data were combined with other data pertaining to the cow (week in milk, parity, mastitis history, somatic cell count category, and season) using Bayes' theorem, which produced a combined probability of the cow having clinical mastitis. If this probability was above a set threshold, the cow was classified as mastitis positive. To illustrate the performance of our method, we used sensor data from 1,003,207 milkings from the University of Florida Dairy Unit collected from 2008 to 2014. Of these, 2,907 milkings were associated with recorded cases of clinical mastitis. Using the DLM/NBC method, we reached an area under the receiver operating characteristic curve of 0.89, with a specificity of 0.81 when the sensitivity was set at 0.80. Specificities with omissions of sensor data ranged from 0.58 to 0.81. These results are comparable to other studies, but differences in data quality, definitions of clinical mastitis, and time windows make comparisons across studies difficult. We found the DLM/NBC method to be a flexible method for combining multiple sensor and nonsensor data sources to predict clinical mastitis and accommodate missing observations. Further research is needed before practical implementation is possible. In particular, the performance of our method needs to be improved in the first 2 wk of lactation. The DLM method produces forecasts that are based on continuously estimated multivariate normal distributions, which makes forecasts and forecast errors easy to interpret, and new sensors can easily be added.

show abstract

Bayesian prediction of bacterial growth temperature range based on genome sequences

Cited by 26 publications

References 24 publications

Machine learning applied to predicting microorganism growth temperatures and enzyme catalytic optima

Machine learning applied to predicting microorganism growth temperatures and enzyme catalytic optima

Building a tRNA thermometer to access the world’s biochemical diversity

Bayesian integration of sensor information and a multivariate dynamic linear model for prediction of dairy cow mastitis

Contact Info

Product

Resources

About