Prediction and characterization of cyclic proteins from sequences in three domains of life

Kedarisetti, Pradyumna; Mizianty, Marcin J.; Kaas, Quentin; Craik, David J.; Kurgan, Lukasz

doi:10.1016/j.bbapap.2013.05.002

Cited by 20 publications

(7 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These have the disadvantage that there is no straightforward statistical approach available to determine likely false discovery rates, but are very valuable in prioritizing a list of peptides for further experimental characterization. Other computational approaches focus more on particular classes of antimicrobial peptides with a strong therapeutic potential, including ribosomal and non-ribosomal cyclic peptides (Prieto et al, 2012; Kedarisetti et al, 2014). While their computational screening methods have the benefit that they focus more strongly on peptides in classes of known therapeutic benefit, we believe that the computational screening approach we identified here complements their approaches, and widens the diversity of peptides for experimental investigation and validation.…”

Section: Discussionmentioning

confidence: 99%

Potential of known and short prokaryotic protein motifs as a basis for novel peptide-based antibacterial therapeutics: a computational survey

et al. 2014

View full text Add to dashboard Cite

Short linear motifs (SLiMs) are functional stretches of protein sequence that are of crucial importance for numerous biological processes by mediating protein–protein interactions. These motifs often comprise peptides of less than 10 amino acids that modulate protein–protein interactions. While well-characterized in eukaryotic intracellular signaling, their role in prokaryotic signaling is less well-understood. We surveyed the distribution of known motifs in prokaryotic extracellular and virulence proteins across a range of bacterial species and conducted searches for novel motifs in virulence proteins. Many known motifs in virulence effector proteins mimic eukaryotic motifs and enable the pathogen to control the intracellular processes of their hosts. Novel motifs were detected by finding those that had evolved independently in three or more unrelated virulence proteins. The search returned several significantly over-represented linear motifs of which some were known motifs and others are novel candidates with potential roles in bacterial pathogenesis. A putative C-terminal G[AG].$ motif found in type IV secretion system proteins was among the most significant detected. A KK$ motif that has been previously identified in a plasminogen-binding protein, was demonstrated to be enriched across a number of adhesion and lipoproteins. While there is some potential to develop peptide drugs against bacterial infection based on bacterial peptides that mimic host components, this could have unwanted effects on host signaling. Thus, novel SLiMs in virulence factors that do not mimic host components but are crucial for bacterial pathogenesis, such as the type IV secretion system, may be more useful to develop as leads for anti-microbial peptides or drugs.

show abstract

Section: Discussionmentioning

confidence: 99%

Potential of known and short prokaryotic protein motifs as a basis for novel peptide-based antibacterial therapeutics: a computational survey

et al. 2014

View full text Add to dashboard Cite

show abstract

“…The discovery of knottins via sequence similarity has produced an extensive and well-organized database, despite a scope limited to sequence similarity [ 25 ]. Cypred [ 43 ] is another relevant software that can predict cyclic proteins and a significant subset of these cyclic peptides have STP like connectivity. While there is no known software to predict non-knotted STPs, there are databases focusing on limited specific families, such as CyBase for cyclotides [ 44 , 45 ], Conoserver for conotoxins [ 46 ] and Arachnoserver for spider toxins [ 47 ], but these have little broad application.…”

Section: Introductionmentioning

confidence: 99%

“…Logic-based machine learning has been used previously to classify the 2D structure of α/α domain type proteins [ 48 ], protein-protein interactions [ 49 ] or functional classifications of proteins from primary sequence. In particular, Support Vector Machines (SVM), a robust class of machine learning approaches [ 50 ], have been successfully used to predict cyclic proteins [ 43 ], 2D and 3D protein structures [ 51 , 52 ] and subcellular localization [ 53 ] from primary sequence.…”

Section: Introductionmentioning

confidence: 99%

PredSTP: a highly accurate SVM based model to predict sequential cystine stabilized peptides

et al. 2015

View full text Add to dashboard Cite

BackgroundNumerous organisms have evolved a wide range of toxic peptides for self-defense and predation. Their effective interstitial and macro-environmental use requires energetic and structural stability. One successful group of these peptides includes a tri-disulfide domain arrangement that offers toxicity and high stability. Sequential tri-disulfide connectivity variants create highly compact disulfide folds capable of withstanding a variety of environmental stresses. Their combination of toxicity and stability make these peptides remarkably valuable for their potential as bio-insecticides, antimicrobial peptides and peptide drug candidates. However, the wide sequence variation, sources and modalities of group members impose serious limitations on our ability to rapidly identify potential members. As a result, there is a need for automated high-throughput member classification approaches that leverage their demonstrated tertiary and functional homology.ResultsWe developed an SVM-based model to predict sequential tri-disulfide peptide (STP) toxins from peptide sequences. One optimized model, called PredSTP, predicted STPs from training set with sensitivity, specificity, precision, accuracy and a Matthews correlation coefficient of 94.86 %, 94.11 %, 84.31 %, 94.30 % and 0.86, respectively, using 200 fold cross validation. The same model outperforms existing prediction approaches in three independent out of sample testsets derived from PDB.ConclusionPredSTP can accurately identify a wide range of cystine stabilized peptide toxins directly from sequences in a species-agnostic fashion. The ability to rapidly filter sequences for potential bioactive peptides can greatly compress the time between peptide identification and testing structural and functional properties for possible antimicrobial and insecticidal candidates. A web interface is freely available to predict STP toxins from http://crick.ecs.baylor.edu/.Electronic supplementary materialThe online version of this article (doi:10.1186/s12859-015-0633-x) contains supplementary material, which is available to authorized users.

show abstract

“…However, generating effective features from protein sequences continues to require enormous manual intervention, and automated approaches have narrowly scoped structure prediction. Chemical property-based feature generation algorithms and dipeptide or tripeptide motif-specific approaches (Chaudhary et al, 2016;Kedarisetti et al, 2014) account for the the majority of these feature generation methods. In particular, Pseudo Amino Acid Composition (PseAAC) has been the most frequently used approach to classify proteins per their functional properties (Xiao et al, 2013;Mohabatkar et al, 2013), subfamilies (Chou, 2005), interactions with other proteins (Jia et al, 2015) and subcellular localizations (Lin et al, 2008).…”

Section: Discussionmentioning

confidence: 99%

Protein classification using modifiedn-gramandskip-grammodels

Islam

Heil

Kearney

et al. 2017

Preprint

View full text Add to dashboard Cite

Motivation:Classification by supervised machine learning greatly facilitates the annotation of protein characteristics from their primary sequence. However, the feature generation step in this process requires detailed knowledge of attributes used to classify the proteins. Lack of this knowledge risks the selection of irrelevant features, resulting in a faulty model. In this study, we introduce a means of automating the work-intensive feature generation step via a Natural Language Processing (NLP)-dependent model, using a modified combination of N-Gram and Skip-Gram models (m-NGSG). Results: A meta-comparison of cross validation accuracy with twelve training datasets from nine different published studies demonstrates a consistent increase in accuracy of m-NGSG when compared to contemporary classification and feature generation models. We expect this model to accelerate the classification of proteins from primary sequence data and increase the accessibility of protein prediction to a broader range of scientists. Availability: m-NGSG is freely available at Bitbucket: https://bitbucket.org/sm islam/mngsg/src Supplements: link to supplementary documents

show abstract

Prediction and characterization of cyclic proteins from sequences in three domains of life

Cited by 20 publications

References 38 publications

Potential of known and short prokaryotic protein motifs as a basis for novel peptide-based antibacterial therapeutics: a computational survey

Potential of known and short prokaryotic protein motifs as a basis for novel peptide-based antibacterial therapeutics: a computational survey

PredSTP: a highly accurate SVM based model to predict sequential cystine stabilized peptides

Protein classification using modifiedn-gramandskip-grammodels

Contact Info

Product

Resources

About