Prediction of Antifungal Activity of Antimicrobial Peptides by Transfer Learning from Protein Pretrained Models

Lobo, Fernando; González, M.; Boto, Alicia; Lastra, José Manuel Pérez de la

doi:10.3390/ijms241210270

Cited by 6 publications

(3 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition to this investigation, predictors for CPP and antifungal properties were also used, to get a better understanding of how the chain length is affecting the biological properties of the peptide. Therefore, an additional four predictors were utilized for CPP prediction, − and three to predict the antifungal effect of the peptides. − According to the NLStradamus, 24 Hst5 contains one NLS, where the length of the NLS depends on the cutoff used in the prediction. However, according to PSORT II, none of the three categories of NLSs are included in this peptide.…”

Section: Resultsmentioning

confidence: 99%

“…Predictions regarding cell-penetrating peptides were performed by the online tools BChemRF-CPPred, using version 2.0, and the FC-3 Feature Composition, MLCPP-2.0, C2Pred, and CellPPD, which were used with the SVM prediction method, and a threshold of 0.0. The antifungal effect was predicted by online tools AntiFP, AFPtranferPred, and Antifungipept …”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Translocation of Antimicrobial Peptides across Model Membranes: The Role of Peptide Chain Length

Skog,

Paracini,

Gerelli

et al. 2024

Mol. Pharmaceutics

View full text Add to dashboard Cite

Cushioned lipid bilayers are structures consisting of a lipid bilayer supported on a solid substrate with an intervening layer of soft material. They offer possibilities for studying the behavior and interactions of biological membranes more accurately under physiological conditions. In this work, we continue our studies of cushion formation induced by histatin 5 ( 24 Hst5), focusing on the effect of the length of the peptide chain. 24 Hst5 is a short, positively charged, intrinsically disordered saliva peptide, and here, both a shorter ( 14 Hst5) and a longer ( 48 Hst5) peptide variant were evaluated. Experimental surface active techniques were combined with coarse-grained Monte Carlo simulations to obtain information about these peptides. Results show that at 10 mM NaCl, both the shorter and the longer peptide variants behave like 24 Hst5 and a cushion below the bilayer is formed. At 150 mM NaCl, however, no interaction is observed for 24 Hst5. On the contrary, a cushion is formed both in the case of 14 Hst5 and 48 Hst5, and in the latter, an additional thick, diffuse, and highly hydrated layer of peptide and lipid molecules is formed, on top of the bilayer. Similar trends were observed from the simulations, which allowed us to hypothesize that positively charged patches of the amino acids lysine and arginine in all three peptides are essential for them to interact with and translocate over the bilayer. We therefore hypothesize that electrostatic interactions are important for the interaction between the solid-supported lipid bilayers and the peptide depending on the linear charge density through the primary sequence and the positively charged patches in the sequence. The understanding of how, why, and when the cushion is formed opens up the possibility for this system to be used in the research and development of new drugs and pharmaceuticals.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Translocation of Antimicrobial Peptides across Model Membranes: The Role of Peptide Chain Length

Skog,

Paracini,

Gerelli

et al. 2024

Mol. Pharmaceutics

View full text Add to dashboard Cite

show abstract

“…A comparative analysis was conducted against existing methods, including antifp 36 AFPtransferPred 37 and PTPAMP 38 revealing superior accuracy of our approach, followed by antifp at 85.71%, while PTPAMP showed a lower accuracy of 46.43%, as detailed in Table 1. These methodologies primarily focus on identification, lacking the capability for quantitative assessment of antifungal activity.…”

Section: Experimental Validationmentioning

confidence: 99%

In Silico Design and Synthesis of Antifungal Peptides Guided by Quantitative Antifungal Activity

Zhang,

Sun,

Zhao

et al. 2024

J. Chem. Inf. Model.

View full text Add to dashboard Cite

Antifungal peptides (AFPs) are emerging as promising candidates for advanced antifungal therapies because of their broad-spectrum efficacy and reduced resistance development. In silico design of AFPs, however, remains challenging, due to the lack of an efficient and well-validated quantitative assessment of antifungal activity. This study introduced an AFP design approach that leverages an innovative quantitative metric, named the antifungal index (AFI), through a three-step process, i.e., segmentation, single-point mutation, and global multipoint optimization. An exhaustive search of 100 putative AFP sequences indicated that random modifications without guidance only have a 5.97–20.24% chance of enhancing antifungal activity. Analysis of the search results revealed that (1) N-terminus truncation is more effective in enhancing antifungal activity than the modifications at the C-terminus or both ends, (2) introducing the amino acids within the 10–60% sequence region that enhance aromaticity and hydrophobicity are more effective in increasing antifungal efficacy, and (3) incorporating alanine, cysteine, and phenylalanine during multiple point mutations has a synergistic effect on enhancing antifungal activity. Subsequently, 28 designed peptides were synthesized and tested against four typical fungal strains. The success rate for developing promising AFPs, with a minimal inhibitory concentration of ≤5.00 μM, was an impressive 82.14%. The predictive and design tool is accessible at .

show abstract

DeepAFP: An effective computational framework for identifying antifungal peptides based on deep learning

Yao,

Zhang,

et al. 2023

Protein Science

View full text Add to dashboard Cite

Fungal infections have become a significant global health issue, affecting millions worldwide. Antifungal peptides (AFPs) have emerged as a promising alternative to conventional antifungal drugs due to their low toxicity and low propensity for inducing resistance. In this study, we developed a deep learning‐based framework called DeepAFP to efficiently identify AFPs. DeepAFP fully leverages and mines composition information, evolutionary information, and physicochemical properties of peptides by employing combined kernels from multiple branches of convolutional neural network (CNN) with bi‐directional long short‐term memory (BiLSTM) layers. In addition, DeepAFP integrates a transfer learning strategy to obtain efficient representations of peptides for improving model performance. DeepAFP demonstrates strong predictive ability on carefully curated datasets, yielding an accuracy of 93.29% and an F1‐score of 93.45% on the DeepAFP‐Main dataset. The experimental results show that DeepAFP outperforms existing AFP prediction tools, achieving state‐of‐the‐art performance. Finally, we provide a downloadable AFP prediction tool to meet the demands of large‐scale prediction and facilitate the usage of our framework by the public or other researchers. Our framework can accurately identify AFPs in a short time without requiring significant human and material resources, and hence can accelerate the development of AFPs as well as contribute to the treatment of fungal infections. Furthermore, our method can provide new perspectives for other biological sequence analysis tasks.This article is protected by copyright. All rights reserved.

show abstract

Prediction of Antifungal Activity of Antimicrobial Peptides by Transfer Learning from Protein Pretrained Models

Cited by 6 publications

References 51 publications

Translocation of Antimicrobial Peptides across Model Membranes: The Role of Peptide Chain Length

Translocation of Antimicrobial Peptides across Model Membranes: The Role of Peptide Chain Length

In Silico Design and Synthesis of Antifungal Peptides Guided by Quantitative Antifungal Activity

DeepAFP: An effective computational framework for identifying antifungal peptides based on deep learning

Contact Info

Product

Resources

About