The DOI matches its Crossref record; 77 of the 77 references it cites were found in a registry or an index; 1 of the 23 sources its claims cite were checked for support.
Large language models generate functional protein sequences across diverse families
Open accessAbstract
Deep-learning language models have shown promise in various biotechnological applications, including protein design and engineering. Here we describe ProGen, a language model that can generate protein sequences with a predictable function across large protein families, akin to generating grammatically and semantically correct natural language sentences on diverse topics. The model was trained on 280 million protein sequences from >19,000 families and is augmented with control tags specifying protein properties. ProGen can be further fine-tuned to curated sequences and tags to improve controllable generation performance of proteins from families with sufficient homologous samples. Artificial proteins fine-tuned to five distinct lysozyme families showed similar catalytic efficiencies as natural lysozymes, with sequence identity to natural proteins as low as 31.4%. ProGen is readily adapted to diverse protein families, as we demonstrate with chorismate mutase and malate dehydrogenase.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
32 Own result · 14 Own interpretation · 11 Not checked yet · 8 Own method · 4 No basis given · 1 Couldn’t confirm · accuracy not yet measured
The cited sources that could be checked couldn’t confirm or dispute this claim.
In the paper
“As in the crystal structure of L056, the predicted artificial structures roughly match known structures found in nature, Fig S11, including for low identity (<40%) artificial sequences.”
§ Large language models generate functional protein sequences across diverse families > Abstract
What the sources say
- Could not verifycites [14] Jumper, J. et al. Highly accurate protein structure prediction with AlphaFold. Nature 596, 583–589 (2021). sourceExperimental — accuracy not yet measured
The Abstract and Main section describe AlphaFold’s structure-prediction accuracy, including cases without a known similar structure. Our copy does not address whether predicted structures of functional artificial sequences resemble natural structures, including at below 40% sequence identity (searched: artificial sequences, L056, and 40% identity). (LLM, conf=0.77)
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Could not verify
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Our protein language model is trained by simply learning to predict the next amino acid for over 280 million protein sequences from thousands of protein families, without explicit biophysical modeling.”
§ Large language models generate functional protein sequences across diverse families > Abstract
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“After training, ProGen can be prompted to generate full-length protein sequences for any protein family from scratch, with a varying degree of similarity to natural proteins.”
§ Large language models generate functional protein sequences across diverse families > Abstract
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
None of the sources cited for this claim have been checked yet.
In the paper
“Our artificial lysozymes span the sequence landscape of natural lysozymes, Fig. 2a, across five families which contain diverse protein folds, active site architectures, and enzymatic mechanisms49,50.”
§ Large language models generate functional protein sequences across diverse families > Abstract
What the sources say
- Not checked yetcites [49] Baase, W. A., Liu, L., Tronrud, D. E. & Matthews, B. W. Lessons from the lysozyme of phage T4. Protein Sci. 19, 631–641 (2010). source
- Not checked yetcites [50] Kuroki, R., Weaver, L. H. & Matthews, B. W. A covalent enzyme–substrate intermediate with saccharide distortion in a mutant T4 lysozyme. Science 262, 2030–2033 (1993). source
plainn = 5- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Rests on an assumption: The five lysozyme families chosen because they lack disulfide bonds can be expressed in high-throughput cell-free systems.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“they demonstrate similar residue position entropies when forming separate multiple sequence alignments of natural and artificial proteins within each family, Fig. 2c.”
§ Large language models generate functional protein sequences across diverse families > Abstract
boostedn = 5- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“In the positive control set of 100 natural proteins, 72% were well-expressed based on chromatography peaks and band visualization.”
§ Large language models generate functional protein sequences across diverse families > Abstract
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Our artificial proteins match activity levels of natural proteins even at lower levels of sequence identity to any known natural protein, (Fig 3b, Fig. S5).”
§ Large language models generate functional protein sequences across diverse families > Abstract
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The representative natural proteins exhibited similar levels of functionality with 59% (53/90) of total proteins considered functional.”
§ Large language models generate functional protein sequences across diverse families > Abstract
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“None of the bmDCA artificial proteins exhibited a detectable level of functionality, Fig. S10”
§ Large language models generate functional protein sequences across diverse families > Abstract
What the sources say
- Not checked yetcites [7] Russ, W. P. et al. An evolution-based model for designing chorismate mutase enzymes. Science 369, 440–445 (2020). source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Generated proteins exhibit similar conservation patterns to natural sequence libraries (Fig. 4ad).”
§ Large language models generate functional protein sequences across diverse families > Abstract
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- It may state two separate claims as one. 29
- Probably not a substantial claim. 16
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“ProGen is a 1.2 billion-parameter neural network trained using a publicly available dataset of 280 million protein sequences.”
§ Large language models generate functional protein sequences across diverse families > Abstract
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“we could readily measure the KM values for the purified apparent lower molecular weight species, where both L056 and L070 were highly active and had comparable Michaelis-Menten parameters to HEWL (Fig. 3e).”
§ Large language models generate functional protein sequences across diverse families > Abstract
hedged- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Among our artificial proteins, 73% (66/90) were functional and exhibited high levels of functionality across families, Fig. 3c.”
§ Large language models generate functional protein sequences across diverse families > Abstract
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Notably a small number of proteins, both within the natural and artificial proteins, were within an order of magnitude of HEWL, which was substantially more active than all negative controls.”
§ Large language models generate functional protein sequences across diverse families > Abstract
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Two proteins, L056 (max ID 69.6%) and L070 (max ID 89.2%) expressed well and incurred bactericidal activities towards the E. coli BL21(DE3) strain used during overnight induction at 16C.”
§ Large language models generate functional protein sequences across diverse families > Abstract
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The ProGen-generated proteins express equally well (72/100 total) across all bins of sequence identity to any known natural protein (maxID 40–90%), Fig 2e.”
§ Large language models generate functional protein sequences across diverse families > Abstract
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Despite having comparable enzymatic activities, L070 and L056 only share 17.9% sequence ID.”
§ Large language models generate functional protein sequences across diverse families > Abstract
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“We also found that there is no bias to location or structural element to the mutations that diverge L056 and L070 from their respective nearest sequence homolog in nature. Differing residues are instead uniformly distributed.”
§ Large language models generate functional protein sequences across diverse families > Abstract
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The KM values of both monomers were too weak to be measured using a heterogeneous, fluorescein-labeled Micrococcus lysodeikticus cell wall substrate (Molecular Probes EnzChek Lysozyme kit); however, both monomers were active using a pseudo-first order kinetic assay (Fig. S7).”
§ Large language models generate functional protein sequences across diverse families > Abstract
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Some mutations are even found within the active site cleft and within regions that influence conformational state (for L056).”
§ Large language models generate functional protein sequences across diverse families > Abstract
plainSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
None of the sources cited for this claim have been checked yet.
In the paper
“We determined a 2.5 Å resolution crystal of L056 (see Methods, Figure 3e). The global fold was similar to predictions, with a Cα RMSD of 2.9 Å from the backbone structure predicted by trRosetta and 2.3 Å RMSD from a WT T4 Lysozyme structure59,60.”
§ Large language models generate functional protein sequences across diverse families > Abstract
What the sources say
plainSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
In the paper
“bmDCA was unable to fit three out of the five lysozyme families, and exhibited 60% detectable expression (30/50 proteins) for the remaining two protein families.”
§ Large language models generate functional protein sequences across diverse families > Abstract
What the sources say
- Not checked yetcites [7] Russ, W. P. et al. An evolution-based model for designing chorismate mutase enzymes. Science 369, 440–445 (2020). source
plainn = 5- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Alignment with a structure featuring a covalently trapped substrate (PDB 148L) reveals that the active site cleft is well formed with the key catalytic residue Glu15 (Glu 11 in T4L) and key substrate binding residue Thr30 (Thr26 in T4L) correctly positioned.”
§ Large language models generate functional protein sequences across diverse families > Abstract
plainSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“These likelihoods showed a significant drop in AUC of 0.18 for CM (p<0.0001, two-tailed test, n=1617) and 0.08 for MDH (p<0.1, two-tailed test, n=56), as compared to fine-tuned ProGen when predicting if an artificial sequence should function.”
§ Large language models generate functional protein sequences across diverse families > Abstract
plaindirectn = 1,617 ·- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
In the paper
“On MDH function data, ProGen log-likelihoods had an AUC of 0.94 (Fig. 4f), which was better than ProteinGAN discriminator scores, with an AUC of 0.87 (p<0.1, two-tailed test, n=56).”
§ Large language models generate functional protein sequences across diverse families > Abstract
What the sources say
- Not checked yetcites [66] Vig, J. et al. BERTology meets biology: Interpreting attention in protein language models. In International Conference on Learning Representations (ICLR, 2020). source
plainn = 56 ·- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
In the paper
“On CM function data, ProGen log-likelihoods had an area under the curve (AUC) of 0.85, significantly better (p<0.0001, two-tailed test, n=1617) than bmDCA, which had an AUC of 0.78 (Fig. 4c).”
§ Large language models generate functional protein sequences across diverse families > Abstract
What the sources say
- Not checked yetcites [7] Russ, W. P. et al. An evolution-based model for designing chorismate mutase enzymes. Science 369, 440–445 (2020). source
plainn = 1,617- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“We purified 6 highly expressed proteins and found that they were all active, but with significantly lower Michaelis-Menten activities than HEWL or the previously generated artificial proteins L056 and L070 (Fig 3f).”
§ Large language models generate functional protein sequences across diverse families > Abstract
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Of the selected sequences, 78 out of the 89 (88%) expressed well and 24 out of the 78 (31%) were soluble (Supplemental Figure 8).”
§ Large language models generate functional protein sequences across diverse families > Abstract
plainn = 2- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“L013 and L038 expressed highly to inclusion bodies.”
§ Large language models generate functional protein sequences across diverse families > Methods > E. coli Expression of Lysozyme Variants
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“For lowID lysozyme A5 the above protocol was altered slightly to accommodate lower catalytic activity of these variants: reaction volumes were increased to 20 μL, plate was covered with an optically transparent seal (Microseal ® ‘B’ seal; BioRad) to mitigate sample evaporation, fluorescent reads were taken every 5 mins for 16 hours with 5 s of linear plate shaking prior to each measurement to min”
§ Large language models generate functional protein sequences across diverse families > Methods > Michaelis-Menten kinetics of lysozyme variants using fluorescein-labeled M. lysodeikticus cell wall
plaindirectSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“The MDH functional labels are binary, so no thresholding was needed to compute AUCs.”
§ Large language models generate functional protein sequences across diverse families > Methods > Evaluating ProGen on Other Protein Systems
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The resultant data were not described by a single exponential model but were described by a double exponential model (Eq. 6), likely due to the heterogeneity of the substrate, and all data were fit in IgorPro7.”
§ Large language models generate functional protein sequences across diverse families > Methods > Lysozyme kcat/KM extrapolation from pseudo first-order kinetic data
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“For induced cultures of L056 and L070, significant turbidity was observed in the spent media after cells were pelleted at 3,500 r.c.f. for 30 mins at 4°C.”
§ Large language models generate functional protein sequences across diverse families > Methods > E. coli Expression of Lysozyme Variants
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Two peaks were resolved for each variant and that harbored lysozyme activity against the fluorescein-labeled M. Lysodeikticus cell wall substrate (EnzChek kit; ThermoFisher).”
- Also stated in: § Purification of L056 and L070 from spent media · plain
“Two peaks were resolved for each variant and that harbored lysozyme activity against the fluorescein-labeled M. Lysodeikticus cell wall substrate (EnzChek kit; ThermoFisher) that corresponded to peaks observed in the first pool purification.”
§ Large language models generate functional protein sequences across diverse families > Methods > Purification of L056 and L070 from spent media
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Also stated in: § Purification of L056 and L070 from spent media · plain
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Significant translational non-crystallographic symmetry and differences with the search model resulted in maps that were initially hard to interpret.”
§ Large language models generate functional protein sequences across diverse families > Methods > Crystallization and structure determination of L056
plaindirectSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“These results indicate that ProGen generates protein sequences that not only can express well but also maintain enzymatic function for diverse sequence landscapes across protein families.”
§ Large language models generate functional protein sequences across diverse families > Abstract
hedgeddirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Rests on an assumption: Well-expression and solubility in the cell-free system indicate that the artificial proteins are structurally well-folded.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Taken together, L056 and L070 harbor potent catalytic activity and bactericidal capabilities that are comparable to HEWL, while diverging from their nearest known natural sequence by 53 and 18 amino acids respectively.”
§ Large language models generate functional protein sequences across diverse families > Abstract
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Rests on an assumption: HEWL is a suitable reference for native lysozyme catalytic activity.
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“The perplexity on a held-out test set was monitored and did not exceed training set perplexity throughout model training.”
§ Large language models generate functional protein sequences across diverse families > Methods > ProGen Training
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“While we do not expect our language model to generate proteins that belong to a completely different distribution or domain (for example creating a new fold that catalyzes an unnatural reaction), it can significantly expand the space of protein sequences from those sampled by evolution.”
§ Large language models generate functional protein sequences across diverse families > Abstract
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Rests on an assumption: The 280 million-sequence training set is diverse enough to learn a universal protein representation transferable across families.
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“The aggregated dataset was split into a training set of size 280 million and two test sets, an out-of-distribution test set (OOD-test) of size 100k from 20 protein families and a randomly sampled in-domain test set (ID-test) of size 1 million, that were held-out for training and used for evaluation.”
§ Large language models generate functional protein sequences across diverse families > Methods > Training Data Curation
plainn = 20- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may state two separate claims as one.
None of the sources cited for this claim have been checked yet.
In the paper
“In sum, ProGen’s model likelihoods are better aligned with experimentally-measured assay data on two diverse protein datasets—chorismate mutase and malate dehydrogenase—than the sequence generation methods from original studies specifically tailored for these families.”
§ Large language models generate functional protein sequences across diverse families > Abstract
What the sources say
- Not checked yetcites [7] Russ, W. P. et al. An evolution-based model for designing chorismate mutase enzymes. Science 369, 440–445 (2020). source
- Not checked yetcites [66] Vig, J. et al. BERTology meets biology: Interpreting attention in protein language models. In International Conference on Learning Representations (ICLR, 2020). source
plaincorrelationaln = 1,617- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“These results indicate that both components of our training strategy—initial training on the universal sequence dataset and fine-tuning on the protein family of interest—contribute significantly to final model performance.”
§ Large language models generate functional protein sequences across diverse families > Abstract
hedgeddirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“For robustness, we trained three discriminators using different random seeds.”
§ Large language models generate functional protein sequences across diverse families > Methods > Lysozyme Sequence Selection
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“Proteins that generated fluorescence one standard deviation above the maximum fluorescence of any negative control were considered functional.”
§ Large language models generate functional protein sequences across diverse families > Abstract
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The resultant data were not described by a single exponential model but were described by a double exponential model (Eq. 6), likely due to the heterogeneity of the substrate, and all data were fit in IgorPro7.”
§ Large language models generate functional protein sequences across diverse families > Methods > Lysozyme kcat/KM extrapolation from pseudo first-order kinetic data
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
- Rests on an assumption: The average molecular weight of the fluorescein-labeled M. lysodeikticus cell wall substrate is unknown and the substrate is heterogeneous.
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“We also ensured that no two generated samples share over 80% identity with each other.”
§ Large language models generate functional protein sequences across diverse families > Abstract
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“After measuring the AUC of each model for each dataset, we used bootstrapping to compute the statistical significance of the difference in AUC of fine-tuned ProGen vs. the reference method (bmDCA and ProGen ablations for CM, ProteinGAN and ProGen ablations for MDH).”
§ Large language models generate functional protein sequences across diverse families > Methods > Evaluating ProGen on Other Protein Systems
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“The fine-tuning dataset was clustered at 80% sequence identity and 10% of the clusters were held-out as a validation set for hyper-parameter optimization and stopping criteria.”
§ Large language models generate functional protein sequences across diverse families > Methods > Lysozyme Generation
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
In the paper
“We also applied thresholding at 0.42 norm relative enrichment (r.e.) to obtain binary labels for CM function, which roughly corresponds to the cut-off point between two modes that exist in CM function data, to be used for ROC curves, following the original study7.”
§ Large language models generate functional protein sequences across diverse families > Methods > Evaluating ProGen on Other Protein Systems
What the sources say
- Not checked yetcites [7] Russ, W. P. et al. An evolution-based model for designing chorismate mutase enzymes. Science 369, 440–445 (2020). source
hedgedcorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“We performed structure prediction without an MSA as input so as to not heavily bias the structure prediction toward a known natural mode.”
§ Large language models generate functional protein sequences across diverse families > Abstract
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
The paper relies on this without support.
The five lysozyme families chosen because they lack disulfide bonds can be expressed in high-throughput cell-free systems.
“Within the lysozyme clan, we chose to work with five families that lack disulfide bonds as to be amenable to high-throughput cell-free expression: Phage lysozyme, Pesticin, Glucosaminidase, Glycoside hydrolase, and Transglycosylase.”
The paper relies on this without support.
Well-expression and solubility in the cell-free system indicate that the artificial proteins are structurally well-folded.
“These results indicate that ProGen can generate artificial proteins that are structurally well-folded for proper expression as compared to a batch of natural proteins”
The paper relies on this without support.
The 280 million-sequence training set is diverse enough to learn a universal protein representation transferable across families.
“Trained in this unsupervised manner from a large, varied protein sequence database, ProGen learns a universal, domain-independent representation of proteins that subsumes local and global structure motifs”
The paper relies on this without support.
HEWL is a suitable reference for native lysozyme catalytic activity.
“Hen egg white lysozyme (HEWL), a naturally-evolved exemplar protein, was measured as positive control”
The paper relies on this without support.
Model likelihoods for GANs are intractable.
“Since model likelihoods for GANs are intractable, we used discriminator scores corresponding to the probability at which the ProteinGAN discriminator predicted each sample was real as a ProteinGAN model score for each MDH sequence.”
The paper relies on this without support.
The average molecular weight of the fluorescein-labeled M. lysodeikticus cell wall substrate is unknown and the substrate is heterogeneous.
“(the average molecular weight of the fluorescein-labeled M. Lysodeikticus cell wall substrate was unknown and likely heterogeneous).”
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“synthetic proteins optimized for specific families of lysozymes using ProGen have exhibited catalytic efficiencies comparable to their naturally occurring counterparts with low sequence identity.”
From Large language models (LLMs): survey, technical frameworks, and future challenges (2024) · cites this paper as Madani et al. 2023
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“AI models have already been used to design new CRISPR enzymes,53 functional proteins,128 and even entire prokaryotic genomes.51”
From How to build the virtual cell with artificial intelligence: Priorities and opportunities (2024) · cites this paper as [128]
The cited source says what the claim attributes to it.
In the citing paper
“The success of LLMs has revolutionized AI learning in natural language and image processing [75] and has been extended into the learning of biological sequence datasets [80].”
From Challenges in AI-driven Biomedical Multimodal Data Fusion and Analysis (2025) · cites this paper as [80]
The cited source says what the claim attributes to it.
In the citing paper
“the ProGen (Madani et al. 2023) language model employs deep learning techniques to produce protein sequences with predefined functionalities.”
From Large language models (LLMs): survey, technical frameworks, and future challenges (2024) · cites this paper as Madani et al. 2023
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Large language models generate functional protein sequences across diverse families > Abstract
“a state-of-the-art transformer-based conditional language model trained only with evolutionary sequence data”
§ Large language models generate functional protein sequences across diverse families > Methods > ProGen Training
“The model was initialized with pretrained weights of CTRL28, which was trained on an English language corpus.”
Citations
Cites
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946
LeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
Highly accurate protein structure prediction with AlphaFold
Jumper, J., Evans, R., Pritzel, A., Green, T., Figurnov, M., Ronneberger, O., Tunyasuvunakool, K., Bates, R., Žídek, A., Potapenko, A., et al. (2021) · Nature · doi:10.1038/s41586-021-03819-2 · Cited by 43,764
Gene Ontology: tool for the unification of biology
Ashburner, M., Ball, C. A., Blake, J. A., Botstein, D., Butler, H., Cherry, J. M., Davis, A. P., Dolinski, K., Dwight, S. S., Eppig, J. T., et al. (2000) · Nature Genetics · doi:10.1038/75556 · Cited by 43,206
Features and development of Coot
Emsley, P., Lohkamp, B., Scott, W. G., & Cowtan, K. (2010) · Acta Crystallographica Section D Biological Crystallography · doi:10.1107/s0907444910007493 · Cited by 29,915
Language Models are Few-Shot Learners
Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al. (2020) · arXiv · doi:10.48550/arxiv.2005.14165 · Cited by 23,372
Phaser crystallographic software
McCoy, A. J., Grosse-Kunstleve, R. W., Adams, P. D., Winn, M. D., Storoni, L. C., & Read, R. J. (2007) · Journal of Applied Crystallography · doi:10.1107/s0021889807021206 · Cited by 20,612
Kabsch, W. (2010) · Acta Crystallographica Section D Biological Crystallography · doi:10.1107/s0907444909047337 · Cited by 16,463
Deep Contextualized Word Representations
Peters, M., Neumann, M., Iyyer, M., Gardner, M., Clark, C., Lee, K., & Zettlemoyer, L. (2018) · Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers) · doi:10.18653/v1/n18-1202 · Cited by 11,983
Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
Raffel, C., Shazeer, N., Roberts, A., Lee, K., Narang, S., Matena, M., Zhou, Y., Li, W., & Liu, P. J. (2019) · arXiv · doi:10.48550/arxiv.1910.10683 · Cited by 10,929
Research works printed but not linked: 13
- Vaswani, A. et al. Attention is all you need. In 31st Conference on Neural Information Processing Systems (NIPS, 2017).
- Devlin, J., Chang, M.-W., Lee, K. & Toutanova, K. BERT: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT, 2019).
- Zellers, R. et al. Defending against neural fake news. In 33rd Conference on Neural Information Processing Systems (NeurIPS, 2019).
- Radford, A., Narasimhan, K., Salimans, T. & Sutskever, I. Improving language understanding by generative pre-training. Preprint at https://s3-us-west-2.amazonaws.com/openai-assets/research-covers/language-unsupervised/language_understanding_paper.pdf (2018).
- Dathathri, S. et al. Plug and play language models: a simple approach to controlled text generation. In 8th International Conference on Learning Representations (ICLR, 2020).
- Pettit, L. D. The IUPAC stability constants database. Chem. Int. 28, 14–15 (2006).
- Bengio, Y., Ducharme, R., Vincent, P. & Janvin, C. A neural probabilistic language model. J. Mach. Learn. Res. 3, 1137–1155 (2003).
- Goyal, K., Dyer, C. & Berg-Kirkpatrick, T. Exposing the implicit energy networks behind masked language models via metropolis–hastings. In 10th International Conference on Learning Representations (ICLR, 2022).
- Alley, E., Khimulya, G., Biswas, S., AlQuraishi, M. & Church, G. M. Unified rational protein engineering with sequence-based deep representation learning. Nat. Methods 16, 1315–1322 (2019). · DOI · Paper page
- Pascanu, R., Mikolov, T. & Bengio, Y. On the difficulty of training recurrent neural networks. In Proc. 30th International Conference on Machine Learning (eds. Dasgupta, S. & McAllester, D.) 1310–1318 (PMLR, 2013).
- Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I. & Salakhutdinov, R. Dropout: a simple way to prevent neural networks from overfitting. J. Mach. Learn. Res. 15, 1929–1958 (2014).
- Holtzman, A., Buys, J., Du, L., Forbes, M. & Choi, Y. The curious case of neural text degeneration. In 8th International Conference on Learning Representations (ICLR, 2020).
- Goodfellow, I. J. et al. Generative adversarial networks. In 28th Conference on Neural Information Processing Systems (NIPS, 2014).
Cited by
Sorted by citations
Items 801–900 of 1,012
Zheng, W., Xu, Y.-F., Hu, Z.-M., Li, K., Xu, Z.-Q., Sun, J.-L., & Wei, J.-F. (2024) · International Immunopharmacology · doi:10.1016/j.intimp.2024.111488 · Cited by 1
Generating Novel and Soluble Class II Fructose-1,6-Bisphosphate Aldolase with ProteinGAN
Tang, F., Ren, M.-Y., Li, X., Lin, Z.-L., & Yang, X.-F. (2023) · Catalysts · doi:10.3390/catal13121457 · Cited by 1
A Vast Navigable Protein Sequence Subspace
Sinai, S., Gane, A., Jain, N., Bryant, D., Bileschi, M., Bashir, A., Church, G. M., Kelsic, E., & Colwell, L. (2026) · doi:10.64898/2026.10.18.759989
“…interpretability of simple models and the precise prediction ability and high flexibility that emerges from training high-capacity models on large datasets (Puszkarska et al. 2024; Hopf et al. 2017; Hsu et al. 2022; Madani et al. 2023; D. Ding et al. 2024; Candido et al. 2026; Park et al. 2024).”
Engineered Cas12a enzymes in biosensing: From molecular mechanisms to clinical translation
Kujur, S., & Prasad, D. (2026) · Applied Materials Today · doi:10.1016/j.apmt.2026.103429
Zou, L., Rashwan, A. K., Fozi, V., Yahia, Z. O., Xie, L., & Chen, W. (2026) · Biotechnology Advances · doi:10.1016/j.biotechadv.2026.109081
Power-Optimal Risk Control via Standardized Margin CalibrationPreprint
Wang, R., Bai, T., Shen, Y., Yu, X., Yang, Y., & Yang, A. Y. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2611.00480
“…prioritizing antibacterial compounds for experimental testing (Stokes et al., 2020), screening generated proteins for desired biological activity (Madani et al., 2023), identifying stable crystal structures (Merchant et al., 2023), selecting patients likely to benefit from treatment (Lei &…”
VenusGenome: Genome-Aware Masked Autoencoding for Microbial Genome Representation LearningPreprint
Zhang, L., Zhang, Y., Tan, Y., Jiang, F., Zhong, B., Cheng, X., Hong, L., Li, M., & Yu, Y. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.10.24.761516
“Protein language models learn transferable molecular representations (Alley et al., 2019; Elnaggar et al., 2022; Rives et al., 2021; Meier et al., 2021; Lin et al., 2023), with alignment-aware, generative, and structure-aware extensions (Rao et al., 2021; Madani et al., 2023; Su et al., 2024b).”
HELICS: Biobank-scale Conditional Synthetic Genome Generation via Latent Flow Matching
Abdel-Azim, A., & Lin, X.-H. (2026) · doi:10.48550/arxiv.2610.36361
Fang, Y.-L., Wan, X.-D., Wang, S.-H., Dai, H.-L., Zeng, H.-L., Chen, N.-Y., & Zhang, X.-X. (2026) · MedComm – Biomaterials and Applications · doi:10.1002/mba2.70084
Liu, B., Liu, H., Jia, Y., He, J., Wang, Y., 姜海燕, Xue, Z., Guo, S., Zhu, Y., Kong, L., et al. (2026) · Marine Drugs · doi:10.3390/md24110375
Fieder, J., Schneider, S., & Fischer, S. (2026) · Critical Reviews in Biotechnology · doi:10.1080/07388551.2026.2738418
Gradient Discrete Schrodinger Bridge Matching for Peptide GenerationPreprint
Benabdallah, A., & Dlala, M. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2610.31361
“The promise is already concrete for proteins—structure prediction [37, 1], language models [60, 46, 52, 31], structure-and sequence-level de-sign [22, 35]—and for antimicrobial peptides, where generative models have produced laboratory-tested candidates [21, 72] and biologically informed generators…”
Decision benchmarking reveals when protein foundation models add experimental valuePreprint
Ding, W., Lu, C., Li, Y., Gong, H., & Wang, Z. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.10.13.758966
Vomo-Donfack, K. L., Aberbache, M., Cantez, A., Hoszu, A., GINOT, G., Doblas, V. G., & Morilla, I. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.10.19.760098
“…of general-purpose protein language models — ESM-1b and ESM-67 2 [21, 22], ProtTrans [23], ProteinBERT [24], and generative counterparts such as ProGen and ProtGPT2 [25, 26] — have been trained by masked or autoregressive modelling over tens of millions of sequences spanning the tree of life.”
Dynamic System Inspired Continuous Large Language Model for AI4Science Tasks
Fan, Z.-X., Zhu, Z., Ma, Y., Wu, F.-G., Yang, J., & Wu, W.-J. (2026) · doi:10.48550/arxiv.2610.26959
Poisoning the corpora of protein design systems during pre-training and retrieval
Koilakos, C., Provatas, K., Patsakis, M., Chan, C. S. Y., Pavlopoulos, G. A., & Georgakopoulos-Soares, I. (2026) · doi:10.48550/arxiv.2610.21200
“Their generative capabilities extend beyond sequence predic-1 tion, with several studies demonstrating the production of novel proteins that retain experimentally measurable biological function 3 .”
InstructPro2: Natural Language and Biological Grounding Guided De Novo Protein DesignPreprint
Novak, C., Song, Z., Li, C., Xie, J., & Li, L. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2610.19592
“Specifically, tag-based approaches leverage functional annotations, such as protein families (Madani et al., 2023), Enzyme Classification (EC) numbers (Munsamy et al., 2022), or species and chain types (Shuai et al., 2023) to guide generation toward targeted protein classes.”
Engineering peroxygenases for pharmaceutical applications: recent advances and future perspectives
Feng, L., Kumar, A., Schwaneberg, U., & Sárria, M. P. (2026) · Chemical Science · doi:10.1039/d6sc04536g
Niazi, S. K., Magoola, M., & Mariam, Z. (2026) · npj Drug Discovery. · doi:10.1038/s44386-026-00084-z
The rise of agentic AI in drug discovery from static tools to autonomous virtual scientists
Zhao, D., Fang, M., Mao, L., Zhu, Y., Su, Q., Shi, H., Xu, H., Li, L., Hou, T., & Hsieh, C.-Y. (2026) · npj Drug Discovery. · doi:10.1038/s44386-026-00075-0
Representing biological concepts as directions in foundation model embeddings
Shamail, A., & McWhite, C. D. (2026) · bioRxiv · doi:10.64898/2026.10.10.758204
“For proteins, these include structure prediction and design (Madani et al. 2023; Lin et al. 2023), homology search (Kaminski et al. 2023; W. Liu et al. 2024), coevolutionary analysis (Z. Zhang et al. 2024), and a broad range of functional prediction tasks (Leclercq and Droit 2026; J.-Y. Chen et al.…”
De novo protein design in an expanded chemical spacePreprint
Li, Y., Jing, Y., Su, Y., Ren, W., Zhang, J., & Liu, T. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.10.08.757794
“Generative models can now produce novel protein sequences and structures and increasingly support the design of functional proteins and protein – protein interactions (1 – 5) .”
Machine Learning Empowers Enzyme Discovery, Structure Prediction, Engineering, and Design
王木强, Jiang, L., Yu, H., Yu, J., Zheng, W., & Wu, J. (2026) · ACS Catalysis · doi:10.1021/acscatal.6c07060
Interpretable Protein-Nucleic Acid Binding Site Prediction via Domain-Guided Multimodal Learning.
Xi, W.-Y., Wang, R.-H., Ye, X.-C., Sakurai, T., & Wei, L.-Y. (2026) · ACS Synthetic Biology · doi:10.1021/acssynbio.6c00340
Designing enzymes for new-to-nature chemistry and non-natural substrates with AlphaProtein NovoPreprint
Wu, Z., Abramson, J., Frerix, T., Chu, A. E., Zhang, R. K., Schulz, L., Danson, A. E., Kwan, T. O. C., Wenliang, L. K., Kelly, J., et al. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.10.01.756017 · Cited by 1
“This suggests that these designs occupy a middle ground between traditional de novo designs and "natural-like" proteins from sequence-based generative models 68–70 .”
Qian, G., Wang, Y., Yan, M., Liu, N., Hou, Z., Wang, H., Pan, S., Qiao, Y., Lv, Y., Li, Y., et al. (2026) · Trends in Food Science & Technology · doi:10.1016/j.tifs.2026.105941
Zhang, Y., Ling, X., Li, Z., Song, L., Jia, Y., Wang, Y., & You, Z. (2026) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2026.115976
Anitha, K., Kothapalli, R. B., Ramachandra, B., Vakkapatla, S., Lalam, R., & Madhav, B. (2026) · Next Materials · doi:10.1016/j.nxmate.2026.103445
Autoregressive Drillhole Modelling Under Distribution Shift
Ding, Y.-H., Su, D. Y., Zhang, Y.-R., Gonzalez, C., & Liu, W. (2026) · doi:10.48550/arxiv.2610.01204
“Autoregressive modelling has expanded beyond language (Achiam et al., 2023; Brown et al., 2020) to scientific sequences such as proteins (Madani et al., 2023), genomes (Nguyen et al., 2024), and general time series (Garza et al., 2023).”
Zhang, Y., & Long, M. (2026) · Foods · doi:10.3390/foods15193538
“In contrast, when the major limitation is large-scale production or formulation rather than protein sequence, further computational redesign may provide limited translational benefit [94,95].”
Artificial Intelligence for Therapeutic Peptide Discovery and Development
Ventura, S. (2026) · Biomolecules · doi:10.3390/biom16101445
Chai, K. F., & Chen, W. N. (2026) · Bioresource Technology Reports · doi:10.1016/j.biteb.2026.103097
The global patent landscape of CRISPR–Cas applications in medicine
Yu, Y., Zhang, Y., Xu, Y., Wen, C., Li, L., & Yu, T. (2026) · Nature Biotechnology · doi:10.1038/s41587-026-03337-4
Beyond prediction: The evolving role of artificial intelligence in genome editing
Shin, S., Uhm, H., Kim, Y.-W., Im, H. W., & Bae, S. (2026) · Molecular Cell · doi:10.1016/j.molcel.2026.08.025
Qureshi, S., & Tayubi, I. (2026) · VAWKUM Transactions on Computer Sciences · doi:10.21015/vtcs.v14i2.2759
MuseDrift: Navigating protein evolutionary manifolds with conditional discrete diffusionPreprint
Wang, C.-Y., & Wang, Y.-Q. (2026) · bioRxiv · doi:10.64898/2026.05.11.724439
“Recent protein language models and diffusion-based generators have advanced de novo and conditional protein design (Madani et al., 2023; Alamdari et al., 2023; Wang et al., 2024).”
GenoTrace: Inheritable Watermarks for Genome Foundation Model Distillation
Yang, G., & Liu, F.-C. (2026) · doi:10.48550/arxiv.2609.35881
Ye, S., Suja, V. C., Li, C., 姜崇明, Zamani, R., Li, X., Bain, C., Zhou, Y., Peterson, W., Wang, H., et al. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2609.30489
“AI foundation models are rapidly becoming practical tools in bioengineering (BE), from protein structure prediction 1,2 and design 3,4 to literature synthesis 5,6 and experimental reasoning 7-9 .”
Protein Language Models: Learning From Evolution, Designing Beyond ItPreprint
Brenner, M., Schlensok, J., Alexander, P., Schmirler, R., & Heinzinger, M. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.09.23.753106
“Steering PLMs is less direct: conditioning uses control tokens (4, 16), inference-time steering provides contextual information such as homologous sequences (39, 40), and alignment adapts models towards experimental or computational objectives (20).”
Wang, Y., Song, L., Wu, Y., Fantke, P., & Li, S. (2026) · Environmental Science & Technology · doi:10.1021/acs.est.6c12624
Samanta, S., Mitra, I., Bhattacharya, A., & Saha, R. (2026) · Molecular Biology · doi:10.1134/s0026893326601199
QLoRA Fine-Tuning of Ministral LLM for Sequence-to-Function Protein Annotation
Pavlyshenko, D., & Pavlyshenko, B. (2026) · doi:10.48550/arxiv.2609.24538
“Generative protein language models such as ProGen [12] and ProtGPT2 [13] model the sequence distribution itself and can generate functional artificial proteins.”
Deng, P.-Q., Wu, Y.-T., Fok, H. K. F., Li, L.-Y., Zhang, W.-B., Sun, F., & Gao, H.-Y. (2026) · Nature Communications · doi:10.1038/s41467-026-77607-9
“De novo protein design opens a path to overcome this limitation and computationally specify and vary the structure and properties of proteins, with recent advances in generative deep learning algorithms demonstrating their powerful capability in creating diverse protein folds 20 – 23 .”
A diffusion model of viral evolution predicts mutation fitness and evolutionary trajectories
Wu, J.-H., Ding, X., & Wu, A.-P. (2026) · bioRxiv · doi:10.64898/2026.09.16.752245
“Autoregressive transformer is one of the most classic generation models used as language models [46, 47].”
When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlowsPreprint
Bénédict, G., Buechler, M., Riera-Solà, G., de Ancos, C., Teukam, Y. G. N., & Freidank, M. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2609.18745
“It sits within a body of generative protein models that produce whole outputs: sequences, whether autoregressive [21, 8], antibody infilling [25], or alignment-conditioned [23, 1], and structure-based backbone design [28, 5], against which the editing formulation outputs a bounded set of…”
Making and Knowing: On the Alleged First Principles of Bioscience and Biotechnology
Huang, K.-C. (2026) · London Journal of Research In Humanities and Social Sciences · doi:10.34257/ljrhss231564uk
Chan, C. S. Y., Karatzikos, A., & Georgakopoulos-Soares, I. (2026) · doi:10.48550/arxiv.2609.16213
“Biological foundation models trained on proteins or genomic sequences can predict molecular properties, evaluate mutations, and generate proteins, genes, or longer genomic sequences 3–6 .”
Enzymatic Electrosynthesis for Sustainable CO 2 Conversion: Advancing Circular Carbon Economy
Xie, Y., Ouyang, J., Suvarna, M., Yuan, X., HILL, M. P., Bornscheuer, U. T., Zhang, H., & Ok, Y. S. (2026) · Carbon Energy · doi:10.1002/cey2.70318
From implicit learning to explicit structural constraints for efficient protein language modelingPreprint
Shen, L., Chao, L., Liu, T., Liu, Q., Zhou, G., Wang, H., Dong, X., Li, T., Zhang, X., & Ni, J. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.03.17.712034
“Most existing PLMs are trained with token-level sequence objectives, including masked-language modeling (MLM) with standard random masking (SRM), as used in ESM-2 [1], ESM-3 [2] and xTrimoPGLM [3], or autoregressive next-token prediction, as used in ProGen [4] and ProtGPT2 [5].”
De novo Rubisco design with protein language models
Kehl, A. J., Chu, S. K. S., Pereira, J. H., Lee, J., Wang, R. Z., Gigl, M., Adams, P., Shih, P. M., & Siegel, J. B. (2026) · bioRxiv · doi:10.64898/2026.09.04.749267
“This strengthens the view that Rubisco performance depends on context-dependent interactions beyond a minimal catalytic motif and is consistent with evidence that PLMs can preserve higher-order evolutionary constraints ( 24 , 27 , 28 ).”
Artificial intelligence-guided engineering of precision ocular therapeutics.
Sayana, R., Wang, S. K., Ting, D., & Guo, L. Y. (2026) · Current Opinion in Ophthalmology · doi:10.1097/icu.0000000000001251
Unsupervised discovery of functional sequence patterns from protein language model with MotifAE
Hou, C., Лю, Д., & Shen, Y. (2026) · Nature Communications · doi:10.1038/s41467-026-77333-2 · Cited by 2
“To address this, several approaches have been proposed to fi ne-tune pLMs to guide the design process toward speci fi c functions or properties 41,42 .”
Hu, X., Liu, X., Wang, T., Liu, X., Yang, G., Huang, P., & Xu, H. (2026) · Biotechnology Advances · doi:10.1016/j.biotechadv.2026.109028
Extremozymes for food fermentation: Integrating AI, metagenomics, and protein engineering.
Ansari, S. K., Shah, N. H., Elboughdiri, N., Chaudhary, A., Ali, M. A. M., & Wani, A. K. (2026) · International Journal of Biological Macromolecules · doi:10.1016/j.ijbiomac.2026.154351
Seo, D.-H., Lee, B.-H., & Yoo, S. (2026) · Comprehensive Reviews in Food Science and Food Safety · doi:10.1111/1541-4337.70656
“PLMs represent a particularly powerful methodology for next-generation enzyme design (Ferruz and Höcker 2022 ; Madani et al. 2023 ).”
Reshaping the scissors: From directed evolution to AI-driven design for enhanced Cas performance
Cheng, X., 阮, 卫., Jin, Z., Suo, L., Ma, Y., Jiang, L., Zheng, W., Li, N., & Hu, W. (2026) · BioDesign Research · doi:10.1016/j.bidere.2026.100124
Khodagholi, Z., & Yousefi, N. (2026) · NAR Genomics and Bioinformatics · doi:10.1093/nargab/lqag099
“ProGen [ 56 ], a 1.2-billion-parameter autoregressive trans-former trained on 280 million protein sequences from over 19 000 families, showed that language models can generate protein sequences with predictable functions.”
InfluentialFassi, E. M. A., Nicolini, M., Saitto, E., Valentini, G., Casiraghi, E., & Grazioso, G. (2026) · Frontiers in Artificial Intelligence · doi:10.3389/frai.2026.1844189
Sun, J., Ma, D., Meng, Q., Bi, C., Li, J., & Shan, A. (2026) · Comprehensive Reviews in Food Science and Food Safety · doi:10.1111/1541-4337.70611
DyAb: sequence-based antibody design and property prediction in a low-data regime
Lin, J., Hofmann, J. L., Leaver-Fay, A., Liang, W.-C., Vasilaki, S., Lee, E., Pinheiro, P. O., Tagasovska, N., Kiefer, J. R., Wu, Y., et al. (2026) · mAbs · doi:10.1080/19420862.2026.2717460
Romero‐Romero, S., Pérez-Villanueva, D., Gonzalez-Vega, L. M., & Beltran, D. M. (2026) · TIES Revista de Tecnología e Innovación en Educación Superior · doi:10.22201/dgtic.26832968e.2026.17.160
“LLMs como ProtGPT2, ProGen, ZymCTRL o PoET-2, se han utilizado en el diseño de nuevas proteínas [49], [35], [38], [39], [50], siendo capaces de aprender patrones evolutivos y estructurales implícitos y, por lo tanto, han podido generar secuencias de novo siguiendo reglas aprendidas.”
Sartori, J., Guimarães, A. C. R., & Machado, L. D. A. (2026) · bioRxiv · doi:10.64898/2026.08.21.746242
“Beyond database prospecting, EC numbers have recently gained prominence as oracles in generative enzyme design pipelines, where they are used to guide protein language models through reinforcement learning strategies toward sequences with desired catalytic functions [13–15].”
Intracellular protein binders for imaging, control and future therapeutics
Barykina, N. V., & Verkhusha, V. V. (2026) · Nature Biomedical Engineering · doi:10.1038/s41551-026-01768-7
Off-Manifold Collapse in Guided Protein Language Models
Zhang, S.-B., Liu, X.-C., Peng, F. Z., Yang, Z.-H., Wu, S.-T., Ma, Y.-Z., & Zhang, J.-W. (2026) · arXiv.org · doi:10.48550/arxiv.2608.18597
“The family spans masked encoders (1; 16), autoregressive generators (3; 17; 4), discrete-diffusion generators (7), instruction-tuned variants (18), and multimodal generative models over sequence, structure, and function (2).”
Artificial Intelligence-Driven Multiscale Design of Protein-Based Materials
Zhou, X., Zhao, C., Zhang, Q., & Yao, Y. (2026) · Bioconjugate Chemistry · doi:10.1021/acs.bioconjchem.6c00309
Learning protein function through autonomous experimental interaction
Brooks, C., Notin, P., & Romero, P. (2026) · bioRxiv · doi:10.64898/2026.08.14.744985
“While highly effective within these observed regimes, they often struggle to engineer proteins with novel or non-natural functions 13,14 .”
Kaijun, Y., Wu, T., Zhang, W., & Li, P. (2026) · Bioinformatics · doi:10.1093/bioinformatics/btag618
“…strings governed by contextual grammar and semantics, similar to human language (Ofer et al. 2021), the language modeling paradigm has been successfully extended to biological macromolecules (Rives et al. 2021), giving rise to Protein Language Models (PLMs) (Jumper et al. 2021, Madani et al. 2023).”
Nandagawli, G. (2026) · Iconic Research and Engineering Journals · doi:10.64388/irev10i2-1722326
“…predictive branches of the literature reviewed in Section 2.2: Constrained generation: a ProGen-style, control-tag-conditioned autoregressive PLM [10] proposes candidate sequences constrained to (i) retain the four cysteine residues forming the two native disulfide bonds, (ii) retain the core…”
InfluentialTask- and dataset-specific information in protein language modelsPreprint
Joeres, R., Senatorov, I. S., Kolchina, A., Klakow, D., & Kalinina, O. V. (2026) · arXiv.org · doi:10.48550/arxiv.2608.12090
“…dataset of 90 artificially generated and 90 natural lysozyme sequences, where the generation was performed with ProGen v1 [23], and their catalytic activity was measured (relative to hen egg white lysozyme, ranging from -0.077 to 2.535 for artificial and from -0.117 to 0.395 for natural ones) [23].”
InfluentialDiscovery and design of potent cell surface display elements
Fang, Z., Saskin, J., Lee, S.-H., Zou, C., Xin, S., Huang, X., Pan, X., Dong, C., Abiri, A., Feng, Y., et al. (2026) · Nature Biotechnology · doi:10.1038/s41587-026-03144-x
Multilayer Network Analysis of Cross-Sector Convergence in the Bio-Industry
Duan, Z., Wang, Z., & Cao, W. (2026) · Mathematics · doi:10.3390/math14162892
Singh, D. (2026) · Journal of Proteins and Proteomics · doi:10.1007/s42485-026-00212-3
“Goles et al. provide the broad AI-peptide landscape; Table 1 narrows that landscape to CATP-relevant endpoints and explicitly records the missing biological link for each architecture (Madani et al. 2023 ).”
Yang, M., Zheng, X., Li, Y., Zheng, Y., Koh, H. Y., Guo, Y., Gao, J., Xia, F., & Pan, S. (2026) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3770855.3818975
Geometry-Preserving Supervised Biological Sequence Design
Sadeghi, E., Lin, I.-H., Deng, X.-Q., Copp, S. M., & Bogdanov, P. (2026) · Knowledge Discovery and Data Mining · doi:10.1145/3770855.3818843
EpiBench: Can LLMs Understand Epitopes for Antibody Drug Discovery?
Wang, Z.-R., Wang, J., Wang, Q.-H., Xu, Y.-Z., Du, G., Hou, T.-J., & Zhang, O. (2026) · arXiv.org · doi:10.48550/arxiv.2608.06022
“ESMFold further demonstrates sequence to structure prediction from language model representations, while autoregressive models such as ProGen and ProGen2 extend language modeling to controllable protein sequence generation [20, 21, 24, 26].”
Makarov, V. A., Sänger, M., Saiz, R., Hartung, M., & Lynch, N. (2026) · Journal of Computer-Aided Molecular Design · doi:10.1007/s10822-026-00894-3
Sun, Y.-D., Song, L.-J., Liu, B., Cheng, C.-Y., Zhang, L., Zhang, J.-P., & Zhang, X.-B. (2026) · The CRISPR Journal · doi:10.1177/25731599261470587
Combining Machine Learning and Directed Evolution for Optimization of a Monooxygenase
Gutierrez, D., Harrison, I. M., Feller, A. L., & Ellington, A. D. (2026) · bioRxiv · doi:10.64898/2026.07.31.742090
Alavi, S. M. A., Borgonovo, F., Pourghazi, F., Tande, A. J., Berbari, E. F., & O’Horo, J. C. (2026) · Clinical Microbiology and Infection · doi:10.1016/j.cmi.2026.07.052
A review of deep learning-based molecular generation
Xu, M., Song, H., & Dong, Y. (2026) · Neurocomputing · doi:10.1016/j.neucom.2026.134779
Early and late adoption and appropriation of LLMs differs by gender, age, and technology education
Draxler, F., Buschek, D., Tavast, M., Kulshrestha, J., & Welsch, R. (2026) · International Journal of Human-Computer Studies · doi:10.1016/j.ijhcs.2026.103922
Wang, Z., Sun, L., Li, Y., Yang, F., & Chang, Y. (2026) · Journal of Pharmaceutical Analysis · doi:10.1016/j.jpha.2026.101754 · Cited by 1
Optimizing protein design through uncertainty-weighted steering of protein language models
Qayyum, A. B. A., Zhou, Y.-T., Qian, X.-N., & Yoon, B.-J. (2026) · Bioinformatics · doi:10.1093/bioinformatics/btag436
“2022, Ferruz et al. 2022, Madani et al .”
Accelerating inference in genomic and proteomic foundation models via speculative decoding
Provatas, K., Karatzikos, A., Koilakos, C., Patsakis, M., Tzanakakis, A., Nayak, A., Pavlopoulos, G. A., Mouratidis, I., Avgoulas, E. I., & Georgakopoulos-Soares, I. (2026) · Bioinform. · doi:10.1093/bioinformatics/btag579
“The same principles of predictive parallelism and multi-Received stage verification are highly relevant for computational biology, where generation of several candidate genomes or proteins is often required to achieve optimal results (Madani et al. 2023, Merchant et al. 2026).”
Beyond Generic Signal Peptides: ApexSP Enables Cargo-Specific Secretion DesignPreprint
Yang, C., Hou, R., Ma, Z., Kang, Q., Zuo, S., Xu, L., Xiao, M., Wu, X., & Jiang, X. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.07.29.741461
“The fidelity of the generated sequences was evaluated using nearest-neighbor similarity 37 .”
Lu, L., Liu, J., Li, Y., & Cai, J. (2026) · Stem Cell Research & Therapy · doi:10.1186/s13287-026-05153-8
“Large-scale protein language models such as ProGen have demonstrated the ability to generate functional protein sequences across diverse protein families [191].”
AI-driven discovery of multifunctional peptides: From sequence space to therapeutics
Liang, Z., Hao, Y., & Chen, J. (2026) · Biotechnology Advances · doi:10.1016/j.biotechadv.2026.108987
Chen, S., Zhu, C., Zhang, Y., Li, Y., Xie, Q., & Wang, H. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2607.23821
“A growing body of work applies LLMs and LLM-driven agents to biomedical and scientific problems [2, 29, 35, 42].”
Deep learning as a bridge between sequencing-derived biological data and functional interpretation
Hong, H., Liu, L., Liao, W., Zhao, J., Liu, Y., He, X., Li, J., He, R., Jin, Y., Sha, L., et al. (2026) · Discover Artificial Intelligence · doi:10.1007/s44163-026-01739-6
“Autoregressive transformers like ProGen [71] and ProtGPT2 [72] sample from learned sequence distributions to generate functional, evolutionarily plausible proteins.”
Explainable AI reveals the allosteric blind spot in protein-ligand binding predictions
Parikh, V., Foley, B., Gatlin, W., Ludwick, M., Turano, L., & Verkhivker, G. M. (2026) · Cell Reports Physical Science · doi:10.1016/j.xcrp.2026.103448 · Cited by 1
ABOPD: Antibody CDR Design via On-Policy DistillationPreprint
Yang, Z., He, J., Xie, J., Wang, D., Qiu, X., Wang, Y., Fu, T., & Wang, B. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2607.18835
“Post-training therefore offers a complementary route to improve existing antibody generators and, more broadly, adapt protein generative models to downstream design objectives, such as specialization to particular protein families [8] and property optimization [9].”
Baskaran, R., Sirikonda, S., Singh, A., Srinivasan, S., Leveridge, B., Casals-Farre, O., Kaur, H., Manivannan, S., Elangovan, K., Quek, C. W. N., et al. (2026) · Frontiers in Medicine · doi:10.3389/fmed.2026.1888765 · Cited by 1
“These GAI utilise neural transformer models that can process vast amounts of data, enabling the creation of complex natural language processing and coherent, constructive output (2, 3).”
De Novo Design of Protein Nanopores: From Minimal Peptides to AI-Driven Design
Vorobieva, A. A., Samanta, R., & Thillo, T. V. (2026) · Chemical Reviews · doi:10.1021/acs.chemrev.5c00990
Huang, Q., Yang, D., Zhou, X., Li, G., Liu, W., Fan, X., Yuan, F., & Chang, X. (2026) · Nature Communications · doi:10.1038/s41467-026-75912-x
“PLMs have recently been utilized to predict the direction of natural evolution 86 , facilitate the design of proteins 87 , improve af fi nities of human antibodies 88 , and develop base editors 21 .”
Zhao, W., Pan, X., Yang, W., LIU, Z., Wei, X., Lin, A., Tang, B., Zhang, L., Xiao, M., Zeng, Q., et al. (2026) · Probiotics and Antimicrobial Proteins · doi:10.1007/s12602-026-11131-6
Post-translational modification-regulating biomaterials for regeneration
Abudousu, A., Tang, Y., Liu, B., Xiang, L., Wang, Z., Li, Y., Zhao, D., & Cui, W. (2026) · Bioactive Materials · doi:10.1016/j.bioactmat.2026.07.020
“Pre-trained protein language models achieve context-dependent prediction of protein structure and function through natural language processing techniques [398].”
Nji, D. N., Dielle, E., Tiencheu, B., Ashu, A. O. E., Flore, D. T. N. E., Denis, S., & Ufuan, A. A. (2026) · Journal of Food and Nutrition Sciences · doi:10.11648/j.jfns.20261404.11
Fei, X., Gu, J., Zhou, C., Liu, Y., Li, Z., & Kang, Y. (2026) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.6c00969
Asymmetric Structural Transfer Between Natural Language and Biological Foundation ModelsPreprint
Wang, L. (2026) · bioRxiv · doi:10.64898/2026.07.09.737502
“Protein and genomic foundation models [17, 18, 19, 20, 21, 22] and structure predictors [23] are typically trained on domain data from scratch, leaving open whether their competence could instead arise from generic sequence structure.”
Hybrid quantum-classical de novo design of MHC-binding peptides
Engdal, E. S., Funk, J., Bacarreza, O., Machado, L., Johansen, K. H., Kemming, J., Farnsworth, T., Brasas, V., Lefèvre-Morand, R. Y. L., Slysz, M., et al. (2026) · bioRxiv · doi:10.64898/2026.07.09.736951
Metadata
- Authors
- Madani, A., Krause, B., Greene, E. R., Subramanian, S., Mohr, B. P., Holton, J. M., Olmos, J. L., Xiong, C., Sun, Z. Z., Socher, R., Fraser, J. S., & Naik, N.
- Affiliations
- Salesforce (United States) (US)
- University of California, San Francisco (US)
- Howard Hughes Medical Institute (US)
- Lawrence Berkeley National Laboratory (US)
- Corresponding author
- Madani, Ali
- Naik, Nikhil
- ORCID
- Venue
- Nature Biotechnology
- Published
- 2023-01-26
- Volume
- 41
- Issue
- 8
- Pages
- 1099-1106
- ISSN
- 0733-222X (ISSN-L)
- 1087-0156
- 1546-1696
- PMC
- PMC10400306
- OpenAlex
- W4318071656
- Cited by
- 1,078
- References
- 80
- Citations by year
- 2026: 314 · 2025: 320 · 2024: 302 · 2023: 141
- FWCI
- 95.97
- Type
- journal article
- Language
- en
- Open access
- green
- www.ncbi.nlm.nih.gov
- Licence
- Locations
- Nature Biotechnology (publishedVersion)
- PubMed (publishedVersion)
- PubMed Central (acceptedVersion)
- eScholarship (California Digital Library) (submittedVersion)
- OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information) (submittedVersion)
- Funders
- U.S. Department of Health & Human Services | National Institutes of Health (GM123159)
- Topics
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Protein Structure and Dynamics
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Machine Learning in Bioinformatics
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Enzyme Catalysis and Immobilization
- Keywords
- large language models, protein sequence generation, protein design, protein families, lysozyme, chorismate mutase, malate dehydrogenase, catalytic efficiency, controllable generation, protein engineering
- MeSH terms
- Amino Acid Sequence, Chorismate Mutase, Estrogens, Conjugated (USP), Language, Proteins
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
A language model drew the claims from the paper on 2026-10-05, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-05 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
