EUVIMEDEUVIMED entdecken
Crossref

Enhancing protein subcellular localization prediction through language model-based knowledge embeddings and machine learning techniques

Karthik Avinash, S Tejas, Sriram Mamidala, Manjunath Vanahalli, Gokul S Krishnan, Sanjay Bankapur

Analytical Biochemistry · 2026-09 · 716 · 116154

DOI: 10.1016/j.ab.2026.116154 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Analytical Biochemistry, 2026-09. Verfasst von Karthik Avinash und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Science and Engineering, Indian Institute of Information Technology Dharwad, India. Electronic address: karthikavinash01@gmail…. Hauptthemen laut MEDLINE: Proteins, Machine Learning, Computational Biology. Die Arbeit nennt 35 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Determining the subcellular localization of proteins is critical for understanding their functional roles. With the rapid expansion of protein sequence databases, traditional experimental and homology-based approaches for Protein Subcellular Localization (PSCL) prediction are becoming increasingly impractical. Although several computational approaches have been developed, achieving high performance in terms of strict accuracy for multi-label PSCL prediction remains a significant challenge. This study evaluates the effectiveness of embeddings from five Protein Language Models (PLMs), including ProtBERT-BFD, ESM-2, ProtALBERT, ProLLaMA, and ProtGPT-2, as input features for various machine learning classifiers. Through five-fold cross-validation on the Swiss-Prot dataset, the results show that encoder-based PLMs, particularly ESM-2, combined with a Support Vector Machine (SVM) employing a polynomial kernel, consistently achieve the best performance. This configuration demonstrates a modest yet consistent improvement in strict-accuracy metrics compared to other model combinations. The proposed work offers a comprehensive evaluation of different PLM architectures and classifier complexities, achieving a strict accuracy of 0.58 while providing valuable insights into their performance trade-offs. This represents a 3 percentage point improvement over existing state-of-the-art models like DeepLoc 2.0. These results suggest that while current embeddings offer strong performance, further advancements in feature extraction and model architectures are needed to significantly boost strict accuracy.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

ProteinsMachine LearningComputational BiologySupport Vector MachinePrediction AlgorithmsPredictive Learning ModelsDatabases, ProteinClassification AlgorithmsLarge Language Models

Stichwörter der Autor:innen

Bidirectional encoder representations from transformersEvolutionary scale modelingMulti-label classificationProtein embeddingProtein language modelsProtein subcellular localizationStrict accuracy

Substanzen

Proteins

Autor:innen und Institutionen

  • Department of Computer Science and Engineering, Indian Institute of Information Technology Dharwad, India. Electronic address: karthikavinash01@gmail.com.
  • Department of Computer Science and Engineering, Indian Institute of Information Technology Dharwad, India. Electronic address: ts692517@gmail.com.
  • Department of Data Science and Intelligent Systems, Indian Institute of Information Technology Dharwad, India. Electronic address: m.sriramsai23@gmail.com.
  • Department of Data Science and Intelligent Systems, Indian Institute of Information Technology Dharwad, India. Electronic address: manjunath.k.vanahalli@gmail.com.
  • Centre for Responsible AI (CeRAI), Indian Institute of Technology Madras, India. Electronic address: gsk1692@gmail.com.
  • Department of Computer Science and Engineering, National Institute of Technology Puducherry, Karaikal, India. Electronic address: sanjaybankapur.mit@gmail.com.

Zitieren

APA

Avinash, K., Tejas, S., Mamidala, S., Vanahalli, M., Krishnan, G. S., & Bankapur, S. (2026). Enhancing protein subcellular localization prediction through language model-based knowledge embeddings and machine learning techniques. Analytical Biochemistry, 716, 116154. https://doi.org/10.1016/j.ab.2026.116154

Vancouver

Avinash K, Tejas S, Mamidala S, Vanahalli M, Krishnan GS, Bankapur S. Enhancing protein subcellular localization prediction through language model-based knowledge embeddings and machine learning techniques. Analytical Biochemistry. 2026;716:116154. doi:10.1016/j.ab.2026.116154

BibTeX
@article{Avinash2026,
  author = {Avinash, Karthik and Tejas, S and Mamidala, Sriram and Vanahalli, Manjunath and Krishnan, Gokul S and Bankapur, Sanjay},
  title = {Enhancing protein subcellular localization prediction through language model-based knowledge embeddings and machine learning techniques},
  journal = {Analytical Biochemistry},
  year = {2026},
  volume = {716},
  pages = {116154},
  doi = {10.1016/j.ab.2026.116154}
}
Kostenlos starten →