Amino Acid Encoding Methods for Protein Sequences: A Comprehensive Review and Assessment

Jing, Xiaoyang; Dong, Qian; Hong, Daocheng; Lu, Ruqian

doi:10.1109/tcbb.2019.2911677

Cited by 63 publications

(50 citation statements)

References 60 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Choosing how to digitally encode amino acids is a crucial step in this context, since it can affect to the overall performance of the models 18 . A comprehensive review and assessment on different amino acid encoding methods 19 shows that position specific scoring matrix (PSSM), an evolution-based position dependent methodology, achieves the best performance on protein secondary structure prediction and protein fold recognition tasks. However, this type of encoding is very consuming computationally 20 and its applicability is limited to proteins with known homologous sequences 19 , which could highly decrease the generalisation capabilities of the predictor for non evolutionary related proteins.…”

Section: Introductionmentioning

confidence: 99%

Effect of sequence padding on the performance of deep learning models in archaeal protein functional prediction

Rio

Martin

Perera-Lluna

et al. 2020

Sci Rep

View full text Add to dashboard Cite

The use of raw amino acid sequences as input for deep learning models for protein functional prediction has gained popularity in recent years. This scheme obliges to manage proteins with different lengths, while deep learning models require same-shape input. To accomplish this, zeros are usually added to each sequence up to a established common length in a process called zero-padding. However, the effect of different padding strategies on model performance and data structure is yet unknown. We propose and implement four novel types of padding the amino acid sequences. Then, we analysed the impact of different ways of padding the amino acid sequences in a hierarchical Enzyme Commission number prediction problem. Results show that padding has an effect on model performance even when there are convolutional layers implied. Contrastingly to most of deep learning works which focus mainly on architectures, this study highlights the relevance of the deemed-of-lowimportance process of padding and raises awareness of the need to refine it for better performance. The code of this analysis is publicly available at https ://githu b.com/b2sla b/paddi ng_bench mark. Since the breakthrough of deep learning (DL) 1 , deep neural networks are being successfully applied in computational biology 2,3. This is due to their capacity for automatically extracting meaningful features from raw data 4. Specifically, DL is useful in the context of biological sequences, such as proteins or RNA, because it can learn directly from the sequence and hence, capture nonlinear dependencies and interaction effects. Some examples of applications of DL on biological sequences include prediction of specifities of DNA and RNA binding proteins 5 , DNA function quantification 6 , de novo peptide design 7 , detection of conserved DNA fragments 8 , prediction of protein associated GO terms 9 or quantification of the impact of genetic variation on gene regulatory mechanisms 3. The specific DL architectures able to leverage the inner structure of sequential biological data are Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN). CNNs entail translational invariance 10 and can be used to find relevant patterns with biological meaning 5,8,11,12. For their part, bidirectional RNNs (and the derived Long Short-Term Memory and Gated Recurrent Units) are appropiate for modelling biological sequences since they are suited for data with a sequential but non-causal structure, variable length, and long-range dependencies 13-16. Both architectures are usually combined, as in DEEPre 17 , where a CNN-RNN model performs a hierarchical classification of enzymes. Proteins are long linear sequences constituted by amino acid residues attached covalently. These amino acid residues are represented by letters that cannot be directly processed by the mathematical operations used by DL models. Choosing how to digitally encode amino acids is a crucial step in this context, since it can affect to the overall performance of the models 18. A comprehensive review and a...

show abstract

Section: Introductionmentioning

confidence: 99%

Effect of sequence padding on the performance of deep learning models in archaeal protein functional prediction

Rio

Martin

Perera-Lluna

et al. 2020

Sci Rep

View full text Add to dashboard Cite

show abstract

“…Yet, in this work, proteins were mapped with these methods, and in general, the results were better than some methods. The best results have not been achieved, and the main reason may be that these methods, including Atchley factors are more effective in prediction of secondary structure of proteins [9]. The performance of binary one-hot seems fair.…”

Section: The Performance Comparison By Using the Blstmmentioning

confidence: 99%

“…In the literature, there are limited methods for converting protein sequences into the numbers. In general, BLOSUM62 (BLOcks SUbstitution Matrix), PAM25 (Point Accepted Mutation), hydrophobicity, EIIP (Electron-Ion Interaction Potential) are applied and the performance of family classification is highly depending on the conversion method [7,9]. Recently, deep learning models are actively used in bioinformatics studies and show promising results.…”

Section: Introductionmentioning

confidence: 99%

A novel Fibonacci hash method for protein family identification by using recurrent neural networks

Alakuş¹,

Türkoğlu²

2021

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

Identification and classification of protein families are one of the most significant problem in bioinformatics and protein studies. It is essential to specify the family of a protein since, proteins are highly used in smart drug therapies, protein functions and in some case, phylogenetic trees. Some sequencing techniques provide researchers to identify the biological similarities of protein families and functions. Yet, determining these families with sequencing applications requires huge amount of time. Thus, it is needed a computer and artificial intelligence based classification system to save the time, and avoid complexity in protein classification process. In order to designate the protein families with computer-aided systems, protein sequences need to be converted to the numerical representations. In this paper, we provide a novel protein mapping method based on Fibonacci numbers and hashing table (FIBHASH). Each amino acid code is assigned to the Fibonacci numbers based on integer representations respectively. Later, these amino acid codes are inserted a hashing table with the size of 20 to be classified with recurrent neural networks. To determine the performance of the proposed mapping method, we used accuracy, f1-score, recall, precision, and AUC evaluation criteria. In addition, the results of evaluation metrics with other protein mapping techniques including EIIP, hydrophobicity, CPNR, Atchley factors, BLOSUM62, PAM250, binary one-hot encoding, and randomly encoded representations are compared. The proposed method showed a promising result with an accuracy of 92.77%, and 0.98 AUC score.

show abstract

“…It mainly includes two key techniques which are feature representation and classifier. For feature representation, if the features of peptide sequences are well-extracted, it will be easier to precisely predict the ACPs (Jing et al, 2019 ). At present, some tools in the prediction of ACPs have been developed.…”

Section: Introductionmentioning

confidence: 99%

EnACP: An Ensemble Learning Model for Identification of Anticancer Peptides

Feng

Jing

et al. 2020

Front. Genet.

Self Cite

View full text Add to dashboard Cite

As cancer remains one of the main threats of human life, developing efficient cancer treatments is urgent. Anticancer peptides, which could overcome the significant side effects and poor results of traditional cancer treatments, have become a new potential alternative these years. However, identifying anticancer peptides by experimental methods is time consuming and resource consuming, it is of great significance to develop effective computational tools to quickly and accurately identify potential anticancer peptides from amino acid sequences. For most current computational methods, feature representation plays a key role in their final successes. This study proposes a novel fast and accurate approach to identify anticancer peptides using diversified feature representations and ensemble learning method. For the feature representations, the information is encoded from multidimensional feature spaces, including sequence composition, sequence-order, physicochemical properties, etc. In order to better model the potential relationships of peptides, multiple ensemble classifiers, LightGBMs, are applied to detect the different feature sets at first. Then the obtained multiple outputs are used as inputs of the support vector machine classifier, which effectively identifies anticancer peptides. Experimental results on cross validation and independent test sets demonstrate that our method can achieve better or comparable performances compared with other state-of-the-art methods.

show abstract

Amino Acid Encoding Methods for Protein Sequences: A Comprehensive Review and Assessment

Cited by 63 publications

References 60 publications

Effect of sequence padding on the performance of deep learning models in archaeal protein functional prediction

Effect of sequence padding on the performance of deep learning models in archaeal protein functional prediction

A novel Fibonacci hash method for protein family identification by using recurrent neural networks

EnACP: An Ensemble Learning Model for Identification of Anticancer Peptides

Contact Info

Product

Resources

About