Prediction of Thermostability from Amino Acid Attributes by Combination of Clustering with Attribute Weighting: A New Vista in Engineering Enzymes

Abstract

The engineering of thermostable enzymes is receiving increased attention. The paper, detergent, and biofuel industries, in particular, seek to use environmentally friendly enzymes instead of toxic chlorine chemicals. Enzymes typically function at temperatures below 60°C and denature if exposed to higher temperatures. In contrast, a small portion of enzymes can withstand higher temperatures as a result of various structural adaptations. Understanding the protein attributes that are involved in this adaptation is the first step toward engineering thermostable enzymes. We employed various supervised and unsupervised machine learning algorithms as well as attribute weighting approaches to find amino acid composition attributes that contribute to enzyme thermostability. Specifically, we compared two groups of enzymes: mesostable and thermostable enzymes. Furthermore, a combination of attribute weighting with supervised and unsupervised clustering algorithms was used for prediction and modelling of protein thermostability from amino acid composition properties. Mining a large number of protein sequences (2090) through a variety of machine learning algorithms, which were based on the analysis of more than 800 amino acid attributes, increased the accuracy of this study. Moreover, these models were successful in predicting thermostability from the primary structure of proteins. The results showed that expectation maximization clustering in combination with uncertainly and correlation attribute weighting algorithms can effectively (100%) classify thermostable and mesostable proteins. Seventy per cent of the weighting methods selected Gln content and frequency of hydrophilic residues as the most important protein attributes. On the dipeptide level, the frequency of Asn-Glu was the key factor in distinguishing mesostable from thermostable enzymes. This study demonstrates the feasibility of predicting thermostability irrespective of sequence similarity and will serve as a basis for engineering thermostable enzymes in the laboratory.

Isolation and in silico functional analysis of MtATP6, a 6-kDa subunit of mitochondrial F1F0-ATP synthase, in response to abiotic stress

“…The analysis of protein properties would help understand the biochemical and physiological role of this subunit. Computational biology can extract a large number of protein properties for an amino acid sequence (Ebrahimi et al, 2011;Tahmasebi et al, 2012). Various properties of the 6-kDa subunit, such as subcellular localization, signal peptide prediction, topology prediction, secondary structure, domains/patterns/motifs prediction, and 3-D structure prediction, were analyzed using the protein databases.…”

Bioinformatics Analysis of Phenylacetaldehyde Synthase (PAAS), a Protein Involved in Flower Scent Production, in Rose

“…Also the frequency of Asn-Gln, Gly-Gly and Asp-Pro were the most important features used to build the rest of tree. This tree was the best model to demonstrate the importance of dipeptides in protein function like thermostability [3].…”

“…However, it is approved that prediction of protein characteristics from the primary amino acid sequence is not possible directly. Therefore, methods to predict protein characteristics have converged on tertiary and quaternary structures [3].…”

“…Under aerobic conditions; phenylacetaldehyde synthase (PAAS) catalyze oxidative decarboxylation by a radical mechanism. PAAS is the first PLP enzyme to be described that, in its native state, catalyzes the stoichiometric oxidative decarboxylation of an L-amino acid substrate ( Figure 1) [1][2][3][4][5][6][7][8][9][10][11][12][13]. PAAS by alignment is homologous with Aromatic L-amino acid decarboxylase (AADC) catalyzes the second enzymatic step in synthesis of the neurotransmitters dopamine and serotonin, which are found in neurons of all animals [4].…”

Amino acid features: a missing compartment of prediction of protein function

“…Interestingly, by developing of different bioinformatics tools, it is possible to calculate a large number of amino acid characteristics from each amino acid sequence. At first, we increased the list of computed amino acid attributes from primary amino acid sequence to more than 800 features [1][2][3][4][5][6] . In addition to frequency/count of different amino acids, .…”