KELM-CPPpred: Kernel Extreme Learning Machine Based Prediction Model for Cell-Penetrating Peptides

Pandey, Poonam; Patel, Vinal; George, Nithin V.; Mallajosyula, Sairam S.

doi:10.1021/acs.jproteome.8b00322

Cited by 66 publications

(41 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We get an accuracy of 90% on Dataset C which is similar to obtained by their RFC and SVM algorithms (Wei, Tang, et al, 2017). With Dataset D, we get an accuracy of 87% which is similar to 86% obtained by their Neural Network algorithm (Pandey et al, 2018). We get an accuracy of 82% on dataset B which was lowest among these datasets.…”

Section: Interpreting Random Forest Predictionssupporting

confidence: 70%

See 1 more Smart Citation

Predicting Cell-Penetrating Peptides: Building and Interpreting Random Forest based prediction Models

Yadahalli

Verma

2020

Preprint

View full text Add to dashboard Cite

Targeting intracellular pathways with peptide drugs is becoming increasingly desirable but often limited in application due to their poor cell permeability. Understanding cellular permeability of peptides remains a major challenge with very little structure-activity relationship known. Fortunately, there exist a class of peptides called Cell-Penetrating Peptides (CPPs), which have the ability to cross cell membranes and are also capable of delivering biologically active cargo into cells. Discovering patterns that make peptides cell-permeable have a variety of applications in drug delivery. In the current study, we build prediction models for CPPs exploring features covering a range of properties based on amino acid sequences, using Random forest classifiers which are often more interpretable than other ensemble machine learning algorithms. While obtaining prediction accuracies of ~96%, we also interpret our prediction models using TreeInterpreter, LIME and SHAP to decipher the contributions of important features and optimal feature space for CPP class. We propose that our work might offer an intuitive guide for incorporating features that impart cell-penetrability into the design of novel CPPs.

show abstract

Section: Interpreting Random Forest Predictionssupporting

confidence: 70%

“…• Dataset D. Dataset from KELM-CPPpred (Pandey et al, 2018) has a total of 826 sequences with an equal number of CPPs and non-CPPs.…”

Section: Dataset Constructionmentioning

confidence: 99%

Predicting Cell-Penetrating Peptides: Building and Interpreting Random Forest based prediction Models

Yadahalli

Verma

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…To further evaluate the proposed method, we performed an independent dataset evaluation prediction model. We compared the independent evaluation experiments with the SkipCPP-Pred [26], CPPred-RF [25], and KELM-CPPpred [22] prediction models, and the independent evaluation dataset was derived from the literature [22]. The experimental results are shown in Table 6.…”

Section: E Performance Of Our Methods On Independent Datasetmentioning

confidence: 99%

Improved Prediction of Cell-Penetrating Peptides via Effective Orchestrating Amino Acid Composition Feature Representation

Cai

et al. 2019

IEEE Access

View full text Add to dashboard Cite

Cell-penetrating peptides (CPPs) promote the transport of pharmacologically active molecules, such as nanoparticles, plasmid DNA and short interfering RNA. Accurate prediction of new CPPs is a prerequisite for in-depth study of such molecules. Biological experimental predictions can provide an accurate description of the penetrating properties of CPPs. However, predicting CPPs by wet laboratory experiments is both resource-intensive and time-consuming. Therefore, the development of effective calculation method prediction has become an important topic in the study of CPPs. Recently, numerous methods developed for predicting CPPs use amino acid composition, alone and the accuracies of such methods have been limited. In this study, we proposed a new CPP prediction framework, which integrates four amino acid composition features, and utilizes these features to help train Support Vector Machine (SVM) model as a classifier to predict CPPs. When performing on the training dataset CPP924, the proposed method achieves an accuracy of 92.3%, which is significantly better than the state-of-the-art methods. These results suggest that the framework can orchestrate various amino acid composition features predicted models flexibly with good performances.

show abstract

“…At first, computational estimation of the cell penetrating properties of the designed peptides was obtained using a CPP-prediction program: KELM-CPPpred. (Pandey et al 2018) Sequences predicted to work as CPPs were then synthesized and tested as presented below. A lipid tail was added to the amino acid backbone to increase the likelihood of cell penetration (Lehto et al 2017).…”

Section: Design Of the Peptidesmentioning

confidence: 99%

Mimicry of Dopamine 1 Receptor Signaling with Cell-Penetrating Peptides

Lorenzón

Gestin

Langel

2020

Int J Pept Res Ther

View full text Add to dashboard Cite

Supplementary figuresSupplementary figure S1 -Cytotoxicity of Dyro 1 on SH-SY5Y cells after 2hrs treatment. SH-SY5Y cells were treated with increasing concentrations of Dyro 1 for 24hrs. The quantity of lactate dehydrogenase (LDH, a cytosolic enzyme) present in the media was measured as an indicator for membrane disruption and cell death. Data were normalized to the controls: untreated cells were used as negative control (physiological release of LDH in the cell medium; 0%), whereas cells treated with Triton X were used as the positive control (being a detergent, Triton X induces membrane disruption and complete LDH release; 100%). Control samples were examined via microscopy before LDH measurements to check for viability. The experiment was performed three times in triplicates. Due to experimental necessities, the examined Dyro 1 concentration range was 10nM -100µM, which was not enough to build a curve. However, comparing the data to the controls, it was still possible to determine the working concentration range (being its maximum, the peptide concentration inducing 20% of membrane rupture at that timepoint). The EC50 was experimentally determined to be at 100µM.Particle size (d.nm) ζ-potential (mV)

show abstract

KELM-CPPpred: Kernel Extreme Learning Machine Based Prediction Model for Cell-Penetrating Peptides

Cited by 66 publications

References 47 publications

Predicting Cell-Penetrating Peptides: Building and Interpreting Random Forest based prediction Models

Predicting Cell-Penetrating Peptides: Building and Interpreting Random Forest based prediction Models

Improved Prediction of Cell-Penetrating Peptides via Effective Orchestrating Amino Acid Composition Feature Representation

Mimicry of Dopamine 1 Receptor Signaling with Cell-Penetrating Peptides

Contact Info

Product

Resources

About