Focus Your Attention: A Focal Attention for Multimodal Learning

2020 ◽  
pp. 1-1
Author(s):  
Chunxiao Liu ◽  
Zhendong Mao ◽  
Tianzhu Zhang ◽  
Anan Liu ◽  
Bin Wang ◽  
...  
Genes ◽  
2021 ◽  
Vol 12 (4) ◽  
pp. 572
Author(s):  
Alan M. Luu ◽  
Jacob R. Leistico ◽  
Tim Miller ◽  
Somang Kim ◽  
Jun S. Song

Understanding the recognition of specific epitopes by cytotoxic T cells is a central problem in immunology. Although predicting binding between peptides and the class I Major Histocompatibility Complex (MHC) has had success, predicting interactions between T cell receptors (TCRs) and MHC class I-peptide complexes (pMHC) remains elusive. This paper utilizes a convolutional neural network model employing deep metric learning and multimodal learning to perform two critical tasks in TCR-epitope binding prediction: identifying the TCRs that bind a given epitope from a TCR repertoire, and identifying the binding epitope of a given TCR from a list of candidate epitopes. Our model can perform both tasks simultaneously and reveals that inconsistent preprocessing of TCR sequences can confound binding prediction. Applying a neural network interpretation method identifies key amino acid sequence patterns and positions within the TCR, important for binding specificity. Contrary to common assumption, known crystal structures of TCR-pMHC complexes show that the predicted salient amino acid positions are not necessarily the closest to the epitopes, implying that physical proximity may not be a good proxy for importance in determining TCR-epitope specificity. Our work thus provides an insight into the learned predictive features of TCR-epitope binding specificity and advances the associated classification tasks.


IEEE Access ◽  
2021 ◽  
pp. 1-1
Author(s):  
Ryosuke Kawamura ◽  
Shizuka Shirai ◽  
Noriko Takemura ◽  
Mehrasa Alizadeh ◽  
Mutlu Cukurova ◽  
...  

2021 ◽  
pp. 073563312110272
Author(s):  
Neila Chettaoui ◽  
Ayman Atia ◽  
Med Salim Bouhlel

Embodied learning pedagogy highlights the interconnections between the brain, body, and the concrete environment. As a teaching method, it provides means of engaging the physical body in multimodal learning experiences to develop the students’ cognitive process. Based on this perspective, several research studies introduced different interaction modalities to support the implementation of an embodied learning environment. One such case is the use of tangible user interfaces and motion-based technologies. This paper evaluates the impacts of motion-based, tangible-based, and multimodal interaction merging between tangible interfaces and motion-based technology on improving students’ learning performance. A controlled study was performed at a primary school with 36 participants (aged 7 to 9), to evaluate the educational potential of embodied interaction modalities compared to tablet-based learning. The results highlighted a significant difference in the learning gains between all groups, as determined by one-way ANOVA [F (3,32) = 6.32, p = .017], in favor of the multimodal learning interface. Findings revealed that a multimodal learning interface supporting richer embodied interaction that took advantage of affording the power of body movements and manipulation of physical objects might improve students’ understanding of abstract concepts in educational contexts.


2001 ◽  
Vol 43 (1) ◽  
pp. 25-40 ◽  
Author(s):  
Piotr Suffczynski ◽  
Stiliyan Kalitzin ◽  
Gert Pfurtscheller ◽  
F.H Lopes da Silva

Sign in / Sign up

Export Citation Format

Share Document