Back

Combinatorial factorizable libraries outperform enumerated and random libraries for antibody discovery

Gifford, D. K.; Dai, Z.

2025-09-28 immunology
10.1101/2025.09.25.678628 bioRxiv
Show abstract

The effective use of high-throughput functional assays to discover novel biological therapeutics requires a diverse library of desirable candidates. Here we design candidate antibody libraries using factorizable neural networks (FNNs) that permit the economic synthesis of factorizable libraries that are depleted of non-specific binders. We experimentally evaluated an FNN designed antibody CDR-H3 factorizable library that is designed to contain binders to peptide-MHC (pMHC) complexes with depletion of library sequences that exhibit non-specific pMHC binding. The factorizable library comprises 104 prefixes that were randomly combined with 104 suffixes. We find that it contains more viable candidates than designed libraries with the same DNA synthesis budget on our pMHC binding task. FNNs make it possible to efficiently computationally design combinatorial libraries, a task that is not possible with conventional methods.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

1
PLOS Computational Biology
● 1863 papers in training set
Top 1%
18.8%
Similar papers in this journal
2
Bioinformatics
◐ 1204 papers in training set
Top 1%
18.8%
Similar papers in this journal
3
Cell Systems
○ 201 papers in training set
Top 0.2%
9.8%
Similar papers in this journal
4
Nature Communications
○ 5641 papers in training set
Top 26%
5.6%
Similar papers in this journal
50% of probability mass above
5
Nature Biotechnology
○ 172 papers in training set
Top 1%
4.1%
Similar papers in this journal
6
eLife
● 5828 papers in training set
Top 30%
4.1%
7
mAbs
○ 32 papers in training set
Top 0.1%
3.6%
8
Bioinformatics Advances
◐ 203 papers in training set
Top 2%
3.3%
9
Proceedings of the National Academy of Sciences
● 2444 papers in training set
Top 15%
3.3%
10
Patterns
○ 78 papers in training set
Top 1%
2.0%
11
Genome Biology
○ 637 papers in training set
Top 5%
1.9%
12
Nature Computational Science
○ 55 papers in training set
Top 0.5%
1.9%
13
Genome Research
● 468 papers in training set
Top 3%
1.8%
14
Cell Reports
○ 1498 papers in training set
Top 20%
1.5%
15
iScience
○ 1154 papers in training set
Top 22%
1.5%
16
Nature Methods
○ 385 papers in training set
Top 5%
1.4%
17
Nature
○ 645 papers in training set
Top 8%
1.1%
18
PLOS ONE
● 5266 papers in training set
Top 52%
1.1%
19
IEEE/ACM Transactions on Computational Biology and Bioinformatics
● 38 papers in training set
Top 0.7%
1.1%
20
BMC Genomics
○ 406 papers in training set
Top 6%
1.1%
21
Nucleic Acids Research
◐ 1281 papers in training set
Top 12%
1.1%
22
Scientific Reports
○ 3612 papers in training set
Top 68%
1.0%
23
NAR Genomics and Bioinformatics
◐ 242 papers in training set
Top 4%
0.9%
24
Frontiers in Immunology
○ 638 papers in training set
Top 10%
0.6%

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.