Reprogramming Pretrained Language Models for Antibody Sequence Infilling

Authors: Igor Melnyk, Vijil Chenthamarakshan, Pin-Yu Chen, Payel Das, Amit Dhurandhar, Inkit Padhi, Devleena Das

ICML 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Results on antibody design benchmarks show that our model on low-resourced antibody sequence dataset provides highly diverse CDR sequences, up to more than a two-fold increase of diversity over the baselines, without losing structural integrity and naturalness.
Researcher Affiliation Collaboration 1IBM Research, Yorktown Heights, NY 10598, USA. 2Georgia Institute of Technology, Atlanta, GA 30332, USA.
Pseudocode No The paper does not contain any pseudocode or clearly labeled algorithm blocks.
Open Source Code Yes Code is available at https://github. com/IBM/Reprog BERT
Open Datasets Yes Structural Antibody Database (Sab Dab) (Dunbar et al., 2013) and Rosetta Antibody Design (Rab D) (Jin et al., 2021), and Co V-Ab Dab dataset (Raybould et al., 2021)
Dataset Splits Yes Table 2. Statistics of the Structural Antibody Database (Sab Dab) for the training, validation and test splits across the three CDRs.
Hardware Specification Yes We trained all models on a single A100 40GB GPU.
Software Dependencies No The paper mentions software like BERT, Alpha Fold, Ig Fold, and Pro Gen, but does not provide specific version numbers for these or any other software dependencies.
Experiment Setup Yes Learning rate 1e-5 Batch size 32 Optimizer Adam