Online Services

●I-TASSER ●I-TASSER-MTD ●C-I-TASSER ●CR-I-TASSER ●QUARK ●C-QUARK ●LOMETS ●MUSTER ●CEthreader ●SEGMER ●DeepFold ●DeepFoldRNA ●FoldDesign ●COFACTOR ●COACH ●MetaGO ●TripletGO ●IonCom ●FG-MD ●ModRefiner ●REMO ●DEMO ●DEMO-EM ●SPRING ●COTH ●Threpp ●PEPPI ●BSpred ●ANGLOR ●EDock ●BSP-SLIM ●SAXSTER ●FUpred ●ThreaDom ●ThreaDomEx ●EvoDesign ●BindProf ●BindProfX ●SSIPe ●GPCR-I-TASSER ●MAGELLAN ●ResQ ●STRUM ●DAMpred

●TM-score ●TM-align ●US-align ●MM-align ●RNA-align ●NW-align ●LS-align ●EDTSurf ●MVP ●MVP-Fit ●SPICKER ●HAAD ●PSSpred ●3DRobot ●MR-REX ●I-TASSER-MR ●SVMSEQ ●NeBcon ●ResPRE ●TripletRes ●DeepPotential ●WDL-RF ●ATPbind ●DockRMSD ●DeepMSA ●FASPR ●EM-Refiner ●GPU-I-TASSER

●BioLiP ●E. coli ●GLASS ●GPCR-HGmod ●GPCR-RD ●GPCR-EXP ●Tara-3D ●TM-fold ●DECOYS ●POTENTIAL ●RW/RWplus ●EvoEF ●HPSF ●THE-DB ●ADDRESS ●Alpaca-Antibody ●CASP7 ●CASP8 ●CASP9 ●CASP10 ●CASP11 ●CASP12 ●CASP13 ●CASP14

Methods

The core of the TCRfinder is a deep learning model trained for accurately scoring the interactions between the antigenic peptides and TCR sequences. Because the CDR3 region of TCR β-chain is the key determinant of the interactions, our model is trained mainly on the β-chain CDR3 regions of the TCR sequences. As outlined in Fig. 1A, TCRfinder first employs a Joint Embedder module (Fig. 1D), which utilizes two pre-trained LMs (Fig. 1B) for peptide and TCRβ CDR3 sequences separately, followed by iterative transformer block learnings starting with concatenated and split embeddings (Fig. 1C). The final TCR-peptide interaction scores are derived through a multi-layer perceptron (MLP) network based on the concatenated representations of TCR and peptide embeddings from the preceding transformer networks (see Fig. 1).

Figure 1. Pipeline of TCRfinder for deep-learning based TCR screening.

Server inputs

TCRfinder can help with two tasks:
• TCR screening: given one peptide seuqence, scoring a set of TCRs (TCR β-chain CDR3 sequences) that can binding to the given peptide.
• Antigen (peptide) screening: given one TCR (TCR β-chain CDR3 sequence), scoring a set of peptides. One example is check the binding score of neoantigen and its wildtype to the given TCR
Here are examples for TCR screening and Peptide screening (Figs. 2-3)
• TCR screening:
  
  

  Figure 2. Server inputs for TCR screening.

  
  

• Antigen screening:
  
  

  Figure 3. Server inputs for antigen screening.

  
  

Server outputs

Here are examples for TCR screening and Peptide screening (Figs. 4-5)
• TCR screening:
  
  

  Figure 4. Server outputs for TCR screening.

  
  

• Antigen screening:
  
  

  Figure 5. Server outputs for antigen screening.

  
  

[back to server]

---

Reference:

• Yang Li, Chaoting Zhang, Xi Zhang, Yang Zhang. TCRfinder: Improved TCR virtual screening for novel antigenic peptides with tailored language models, submitted (2024).