skip to main content

Explore Research Products in the NSF-PAR

Title: Cross-Modality Protein Embedding for Compound-Protein Affinity and Contact Prediction
Compound-protein pairs dominate FDA-approved drug-target pairs and the prediction of compound-protein affinity and contact (CPAC) could help accelerate drug discovery. In this study we consider proteins as multi-modal data including 1D amino-acid sequences and (sequence-predicted) 2D residue-pair contact maps. We empirically evaluate the embeddings of the two single modalities in their accuracyand generalizability of CPAC prediction (i.e. structure-free interpretable compound-protein affinity prediction). And we rationalize their performances in both challenges of embedding individual modalities and learning generalizable embedding-label relationship. We further propose two models involving cross-modality protein embedding and establish that the one with cross interaction (thus capturing correlations among modalities) outperforms SOTAs and our single modality models in affinity, contact, and binding-site predictions for proteins never seen in the training set.  more » « less
Award ID(s):
1943008
NSF-PAR ID:
10225270
Author(s) / Creator(s):
;
Date Published:
Journal Name:
Machine Learning for Structure Biology (MLSB) Workshop at the 34th Conference on Neural Information Processing Systems
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ( , Bioinformatics)
    Abstract Motivation

    Computational methods for compound–protein affinity and contact (CPAC) prediction aim at facilitating rational drug discovery by simultaneous prediction of the strength and the pattern of compound–protein interactions. Although the desired outputs are highly structure-dependent, the lack of protein structures often makes structure-free methods rely on protein sequence inputs alone. The scarcity of compound–protein pairs with affinity and contact labels further limits the accuracy and the generalizability of CPAC models.

     Results

    To overcome the aforementioned challenges of structure naivety and labeled-data scarcity, we introduce cross-modality and self-supervised learning, respectively, for structure-aware and task-relevant protein embedding. Specifically, protein data are available in both modalities of 1D amino-acid sequences and predicted 2D contact maps that are separately embedded with recurrent and graph neural networks, respectively, as well as jointly embedded with two cross-modality schemes. Furthermore, both protein modalities are pre-trained under various self-supervised learning strategies, by leveraging massive amount of unlabeled protein data. Our results indicate that individual protein modalities differ in their strengths of predicting affinities or contacts. Proper cross-modality protein embedding combined with self-supervised learning improves model generalizability when predicting both affinities and contacts for unseen proteins.

     Availability and implementation

    Data and source codes are available at https://github.com/Shen-Lab/CPAC.

     Supplementary information

    Supplementary data are available at Bioinformatics online.

     
    more » « less
  2. ; ; ; ; ( , Machine Learning in Structural Biology Workshop at the 36th Conference on Neural Information Processing Systems)
    Approaches to in silico prediction of protein structures have been revolutionized by AlphaFold2, while those to predict interfaces between proteins are relatively underdeveloped, owing to the overly complicated yet relatively limited data of protein–protein complexes. In short, proteins are 1D sequences of amino acids folding into 3D structures, and interact to form assemblies to function. We believe that such intricate scenarios are better modeled with additional indicative information that reflects their multi-modality nature and multi-scale functionality. To improve binary prediction of inter-protein residue-residue contacts, we propose to augment input features with multi-modal representations and to synergize the objective with auxiliary predictive tasks. (i) We first progressively add three protein modalities into models: protein sequences, sequences with evolutionary information, and structure-aware intra-protein residue contact maps. We observe that utilizing all data modalities delivers the best prediction precision. Analysis reveals that evolutionary and structural information benefit predictions on the difficult and rigid protein complexes, respectively, assessed by the resemblance to native residue contacts in bound complex structures. (ii) We next introduce three auxiliary tasks via self-supervised pre-training (binary prediction of protein-protein interaction (PPI)) and multi-task learning (prediction of inter-protein residue–residue distances and angles). Although PPI prediction is reported to benefit from predicting intercontacts (as causal interpretations), it is not found vice versa in our study. Similarly, the finer-grained distance and angle predictions did not appear to uniformly improve contact prediction either. This again reflects the high complexity of protein–protein complex data, for which designing and incorporating synergistic auxiliary tasks remains challenging. 
    more » « less
  3. ; ; ; ; ; ; ( , Briefings in Bioinformatics)
    Abstract

    Protein fold recognition is a critical step toward protein structure and function prediction, aiming at providing the most likely fold type of the query protein. In recent years, the development of deep learning (DL) technique has led to massive advances in this important field, and accordingly, the sensitivity of protein fold recognition has been dramatically improved. Most DL-based methods take an intermediate bottleneck layer as the feature representation of proteins with new fold types. However, this strategy is indirect, inefficient and conditional on the hypothesis that the bottleneck layer’s representation is assumed as a good representation of proteins with new fold types. To address the above problem, in this work, we develop a new computational framework by combining triplet network and ensemble DL. We first train a DL-based model, termed FoldNet, which employs triplet loss to train the deep convolutional network. FoldNet directly optimizes the protein fold embedding itself, making the proteins with the same fold types be closer to each other than those with different fold types in the new protein embedding space. Subsequently, using the trained FoldNet, we implement a new residue–residue contact-assisted predictor, termed FoldTR, which improves protein fold recognition. Furthermore, we propose a new ensemble DL method, termed FSD_XGBoost, which combines protein fold embedding with the other two discriminative fold-specific features extracted by two DL-based methods SSAfold and DeepFR. The Top 1 sensitivity of FSD_XGBoost increases to 74.8% at the fold level, which is ~9% higher than that of the state-of-the-art method. Together, the results suggest that fold-specific features extracted by different DL methods complement with each other, and their combination can further improve fold recognition at the fold level. The implemented web server of FoldTR and benchmark datasets are publicly available at http://csbio.njust.edu.cn/bioinf/foldtr/.

     
    more » « less
  4. ; ; ; ; ; ; ; ; ; ( , Nature Communications)
    Abstract Predicting assay results for compounds virtually using chemical structures and phenotypic profiles has the potential to reduce the time and resources of screens for drug discovery. Here, we evaluate the relative strength of three high-throughput data sources—chemical structures, imaging (Cell Painting), and gene-expression profiles (L1000)—to predict compound bioactivity using a historical collection of 16,170 compounds tested in 270 assays for a total of 585,439 readouts. All three data modalities can predict compound activity for 6–10% of assays, and in combination they predict 21% of assays with high accuracy, which is a 2 to 3 times higher success rate than using a single modality alone. In practice, the accuracy of predictors could be lower and still be useful, increasing the assays that can be predicted from 37% with chemical structures alone up to 64% when combined with phenotypic data. Our study shows that unbiased phenotypic profiling can be leveraged to enhance compound bioactivity prediction to accelerate the early stages of the drug-discovery process. 
    more » « less
  5. ; ; ; ( , Bioinformatics)
    Abstract Motivation

    Accurately predicting the likelihood of interaction between two objects (compound–protein sequence, user–item, author–paper, etc.) is a fundamental problem in Computer Science. Current deep-learning models rely on learning accurate representations of the interacting objects. Importantly, relationships between the interacting objects, or features of the interaction, offer an opportunity to partition the data to create multi-views of the interacting objects. The resulting congruent and non-congruent views can then be exploited via contrastive learning techniques to learn enhanced representations of the objects.

     Results

    We present a novel method, Contrastive Stratification for Interaction Prediction (CSI), to stratify (partition) a dataset in a manner that can be exploited via Contrastive Multiview Coding to learn embeddings that maximize the mutual information across congruent data views. CSI assigns a key and multiple views to each data point, where data partitions under a particular key form congruent views of the data. We showcase the effectiveness of CSI by applying it to the compound–protein sequence interaction prediction problem, a pressing problem whose solution promises to expedite drug delivery (drug–protein interaction prediction), metabolic engineering, and synthetic biology (compound–enzyme interaction prediction) applications. Comparing CSI with a baseline model that does not utilize data stratification and contrastive learning, and show gains in average precision ranging from 13.7% to 39% using compounds and sequences as keys across multiple drug–target and enzymatic datasets, and gains ranging from 16.9% to 63% using reaction features as keys across enzymatic datasets.

     Availability and implementation

    Code and dataset available at https://github.com/HassounLab/CSI.

     
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email