Biological Sequence Modeling with Convolutional Kernel Networks

By Ali Pazokitoroudi, Yue Wu, Kathryn S. Burch, Kangcheng Hou, Bogdan Pasaniuc, Sriram Sankararaman. 2017

178 citations recorded by the source corpus.

The growing number of annotated biological sequences available makes it possible to learn genotype-phenotype relationships from data with increasingly high accuracy. When large quantities of labeled samples are available for training a model, convolutional neural networks can be used to predict the phenotype of unannotated sequences with good accuracy. Unfortunately, their performance with medium- or small-scale datasets is mitigated, which requires inventing new data-efficient approaches. In this paper, we introduce a hybrid approach between convolutional neural networks and kernel methods to model biological sequences. Our method enjoys the ability of convolutional neural networks to learn data representations that are adapted to a specific task, while the kernel point of view yields algorithms that perform significantly better when the amount of training data is small. We illustrate these advantages for transcription factor binding prediction and protein homology detection, and we demonstrate that our model is also simple to interpret, which is crucial for discovering predictive motifs in sequences. The source code is freely available at https://gitlab.inria.fr/dchen/CKN-seq.

Subfields: Molecular Biology, Model Architectures, Computational Biology & Bioinformatics

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Molecular Biology

1. Xiaogen Zhou
2. Jeffrey J. Gray
3. Gui-jun Zhang
4. Yang Zhang
5. Sai Pooja Mahajan
6. Dongbo Bu
7. Jianwei Zhu
8. John A. Capra

In Model Architectures

1. John A. Capra
2. Mark Gerstein
3. Yucheng T. Yang
4. Tie-Yan Liu
5. Yang Zhang
6. Bian Li
7. Fusong Ju
8. Tao Qin

In Computational Biology & Bioinformatics

1. Xiaogen Zhou
2. Gui-jun Zhang
3. Jeffrey J. Gray
4. Jianwei Zhu
5. Dongbo Bu
6. Sai Pooja Mahajan
7. John A. Capra
8. Mark Gerstein

Citations in the corpus

• CopulaNet: Learning residue co-evolution directly from multiple sequence alignment for protein structure prediction (2020, methodology): “For example, both AlphaFold [6] and RaptorX [16] rely on the inter-residue contacts predicted by CCMpred, a DCA-based approach using the Potts model [17].”
• Deep Learning in Protein Structural Modeling and Design (2020, methodology): “Later, the method was extended to predict Ca Ca, Ca Cg, Cg Cg, N-O distances and torsion angles (DL-based RaptorXAngle),(163) giving constraints to locate side chains and additionally constrain the backbone; all five distances, torsions, and secondary structure predictions were converted to constrai”
• MMpred: a distance-assisted multimodal conformation sampling for de novo protein structure prediction (2021, methodology): “RaptorX inputs the predicted distance, secondary structure, and torsion angles predicted by deep convolutional neural networks into the Crystallography and NMR system (CNS) (Brunger et al., 2007) to quickly constructed an accurate 3D model (Xu, 2019; Wang et al., 2017).”
• Predicting changes in protein thermodynamic stability upon point mutation with deep 3D convolutional neural networks (2020, background): “They have recently proven to perform well in residue contact prediction (57585 60) and protein tertiary structure prediction (61-64).”