Accelerating bioactive peptide discovery via mutual information-based meta-learning

Abstract

Recently, machine learning methods have been developed to identify various peptide bio-activities. However, due to the lack of experimentally validated peptides, machine learning methods cannot provide a sufficiently trained model, easily resulting in poor generalizability. Furthermore, there is no generic computational framework to predict the bioactivities of different peptides. Thus, a natural question is whether we can use limited samples to build an effective predictive model for different kinds of peptides. To address this question, we propose Mutual Information Maximization Meta-Learning (MIMML), a novel meta-learning-based predictive model for bioactive peptide discovery. Using few samples from various functional peptides, MIMML can sufficiently learn the discriminative information amongst various functions and characterize functional differences. Experimental results show excellent performance of MIMML though using far fewer training samples as compared to the state-of-the-art methods. We also decipher the latent relationships among different kinds of functions to understand what meta-model learned to improve a specific task. In summary, this study is a pioneering work in the field of functional peptide mining and provides the first-of-its-kind solution for few-sample learning problems in biological sequence analysis, accelerating the new functional peptide discovery. The source codes and datasets are available on https://github.com/TearsWaiting/MIMML.

Identify Bitter Peptides by Using Deep Representation Learning Features

“…We compared the predictive performance of iBitter-DRLF with existing methods, including iBitter-Fuse [ 18 ], MIMML [ 20 ], iBitter-SCM [ 17 ], and BERT4Bitter [ 19 ] to assess the effectiveness and utility of our method against its competitors. Independent test results for iBitter-DRLF and the existing methods are compared in Table 4 .…”

“…In 2021, BERT4Bitter [ 19 ] was proposed, which used natural language processing (NLP) heuristic signature coding methods to represent peptide sequences as feature descriptors, and it displayed better accuracy. While in 2022, He et al [ 20 ] proposed mutual information-based meta learning (MIMML) to discover the best feature combination for bitter peptides and attained an independent accuracy of 93.8%. Although great progress has been made in this field, there is still much room for improvement in the performance of machine learning-based bitter peptide identification models using sequences only.…”

Assessing sequence-based protein–protein interaction predictors for use in therapeutic peptide engineering

“…CAMP is a novel deep learning method that has shown promise in assessing not only protein–peptide interaction probabilities, but also in identifying the peptide residues that underpin the interaction using convolutional networks and a self-attention mechanism 34 . Mutual Information Maximization Meta-Learning is another approach relying on meta-learning and information theory that optimizes peptide bioactivity 35 . In fact, the growing interest in computational therapeutic peptide screening may require us to regularly benchmark emerging methods.…”

Applications of deep learning in understanding gene regulation

“…The authors discovered that using ∼20% of data from each cell type to pre-train a meta-model and then adapt it to a specific cell type using the rest of the data benefited model performance. MIMML 165 is a newly proposed meta-learning framework for bioactive peptide function prediction. MIMML is based on the Prototypical Network, 168 which performs few-shot classification by measuring the distance from a query example to a few exemplars of each class.…”