English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Current antibody language models are limited by their use of unpairedantibody sequence data and the biases in publicly available antibody sequencedatasets, which are skewed toward antibodies against a relatively small numberof pathogens. A recently published dataset (by Jaffe, et al) of approximately1.6 x 10^6 natively paired human antibody sequences from healthy donorsrepresents by far the largest dataset of its kind and offers a uniqueopportunity to evaluate how antibody language models can be improved bytraining with natively paired antibody sequence data. We trained two BaselineAntibody Language Models (BALM), using natively paired (BALM-paired) orunpaired (BALM-unpaired) sequences from the Jaffe dataset. We provide evidencethat training with natively paired sequences substantially improves modelperformance and that this improvement results from the model learningimmunologically relevant features that span the light and heavy chains. We alsoshow that ESM-2, a state-of-the-art general protein language model, learnssimilar cross-chain features when fine-tuned with natively paired antibodysequence data.

Improving antibody language models with native pairing