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 601–700 of 1,012
Protein Language Models Diverge from Natural Language: Comparative Analysis and Improved InferencePreprint
Hart, A., Han, C., Kim, J., Zhao, H.-M., & Ji, H. (2026) · arXiv.org · doi:10.48550/arxiv.2602.20449 · Cited by 3
“Future work could study decoder models such as ProGen (Madani et al., 2023) or multimodal models such as (Hayes et al., 2025) to better understand how models generate novel proteins or how various types of data, such as protein sequence and structural data, are handled by a model.”
Beyond SMILES: Evaluating Agentic Systems for Drug DiscoveryPreprint
Wijaya, E. (2026) · arXiv.org · doi:10.48550/arxiv.2602.10163 · Cited by 3
“ProtBERT [11], ESM-2 [10], and ProGen [19] encode evolutionary and structural priors from millions of protein sequences.”
Mapping AAV capsid sequences to functions through function-guided in silico evolution
Guo, B.-J., Zheng, H., Lin, X., Jiang, H., Mo, A.-S., Wei, H.-Y., Zhang, R., Yuan, Y.-B., Wu, Y., Li, H., et al. (2026) · bioRxiv · doi:10.1101/2024.10.11.617764 · Cited by 3
Wei, S., Zhang, J., Chen, Y., Tang, C., & Tan, J. (2026) · Bioinformatics · doi:10.1093/bioinformatics/btag042 · Cited by 3
“Transformer language models such as ESM-1b (Rives et al. 2021) and ProGen (Madani et al. 2023) demonstrate sequence-to-function inference capabilities through unsupervised learning of residue dependencies in large-scale datasets.”
AI Guided Protein Design for Next‐Generation Autogenic Engineered Living Materials
Hammad, H. M., & Duraj‐Thatte, A. (2026) · Advanced Intelligent Systems · doi:10.1002/aisy.202501021 · Cited by 4
Wu, F., Wu, R., Chen, L., Chen, Q., & Liu, H. (2026) · New Biotechnology · doi:10.1016/j.nbt.2026.01.009 · Cited by 3
Accurate enzyme specificity constant prediction with iESC
Zhang, Y., Liu, L.-H., Wang, S., & Jiang, A. (2026) · Bioresource Technology · doi:10.1016/j.biortech.2026.134067 · Cited by 3
ProFam: Open-Source Protein Family Language Modelling for Fitness Prediction and DesignPreprint
Wells, J., Hawkins‐Hooker, A., Livne, M., Lin, W., Miller, D. M., Dallago, C., Bordin, N., Paige, B., Rost, B., Orengo, C. A., et al. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2025.12.19.695431 · Cited by 3
“Early strategies for steering generative models relied on explicit functional labels, such as enzyme commission numbers [13], or fine-tuning on specific protein families [14].”
Karim, W., Alvi, R., Zaman, S. B., Abian, A. I., Raiaan, M. A. K., Mukta, S. H., Rashid, M. R. U., Islam, R., Sebastian, Y., & Azam, S. (2026) · Archives of Computational Methods in Engineering · doi:10.1007/s11831-026-10743-z · Cited by 7
“Autoregressive Transformer-based models, such as Pro-Gen [47] and ProGen2 [48], have advanced domain specialization in the design of generative protein sequences.”
InfluentialB‐EPIC: A Transformer‐Based Language Model for Decoding B Cell Immunodominance Patterns
Liang, J.-Z., Wang, Y., Sun, C., Liu, T., Wu, Z.-F., Chen, L.-P., Chen, L.-N., Li, P.-L., Li, Z.-K., Zhang, C.-G., et al. (2025) · Advancement of science · doi:10.1002/advs.202508896 · Cited by 3
Debenedetti, P., de Pablo, J. D., & Schatz, G. (2025) · Proceedings of the National Academy of Sciences of the United States of America · doi:10.1073/pnas.2517062122 · Cited by 3
Towards Understanding the Shape of Representations in Protein Language Models
Beshkov, K., & Malthe-Sorenssen, A. (2025) · International Conference on Learning Representations · doi:10.48550/arxiv.2509.24895 · Cited by 3
“Using PLMs, one can predict the folding of proteins in 3D space (Lin et al., 2023), generate new candidate sequences for viral vectors (Lyu et al., 2024), enzymes (Madani et al., 2023), biosensors (Hayes et al., 2025) and functional binders (Chen et al., 2025; Bryant & Elofsson, 2023).”
SpecMER: Fast Protein Generation with K-mer Guided Speculative Decoding
Walton, T., Tsui, D., Musharaf, A., & Aghazadeh, A. (2025) · Neural Information Processing Systems · doi:10.48550/arxiv.2509.21689 · Cited by 3
“Instead, decoding with nucleus sampling (top-p ) is common to balance plausibility and diversity, and is utilized with both ProGen and ProGen2 [4, 8, 16, 17].”
World’s first AI-designed viruses a step towards AI-generated life
Kavanagh, K. (2025) · Nature · doi:10.1038/d41586-025-03055-y · Cited by 3
Uncovering Microbial Biosynthetic Potential with Genomic Context-aware Protein Language ModelPreprint
Kang, Z.-X., Zhang, H.-H., Liang, C., Yang, R., Ye, Y., Bai, H., Zhang, Y.-H., & Ning, K. (2025) · bioRxiv · doi:10.1101/2025.04.29.651206 · Cited by 3
TCR representation learning with protein language models: a comprehensive review
Kinoshita, K., & Kobayashi, T. J. (2025) · International Immunology · doi:10.1093/intimm/dxaf048 · Cited by 3
“Notable examples include UniRep (37), TAPE (38), ESM2 (7), ProteinBERT (6), and ProGen (8), which are pre-trained on extensive protein databases like UniRef (39) and Pfam (40).”
Approaches for regulating enzyme activities: Recent advances in experiment and computation
Cui, Q. (2025) · Current Opinion in Structural Biology · doi:10.1016/j.sbi.2025.103124 · Cited by 3
Yao, L.-T., Wang, F., Xie, P.-L., Guan, J.-H., Zhao, Z.-H., He, X.-X., Liu, X.-C., Chiang, Y.-C., & Lee, T.-Y. (2025) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.5c00860 · Cited by 3
A six-tiered framework for evaluating AI models from repeatability to replaceability
Tian, S., Lam, A. W. Y., Sung, J. J. �., & Goh, W. W. B. (2025) · Trends in biotechnology · doi:10.1016/j.tibtech.2025.07.015 · Cited by 4
Zhang, S., Jiang, Z., Huang, R., Huang, W., Peng, S., Mo, S., Zhu, L., Li, P., Zhang, Z., Pan, E., et al. (2025) · Advanced Science · doi:10.1002/advs.202502723 · Cited by 10
AI in drug design: evolution or revolution?
Green, D. V. S. (2025) · Emerging Topics in Life Sciences · doi:10.1042/etls20240005 · Cited by 3
“…[45] and multi-modal [46] modelling of complex data sets • Small [47] or large language models trained on chemistry [48] and protein sequence [49] • Transfer learning from pre-trained/foundational models for small data sets [50,51] • Federated modelling for safe data sharing between…”
Transformers in Protein: A SurveyPreprint
Ling, X., Li, Z., Wang, Y., & You, Z. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2505.20098 · Cited by 6
“Applications of ProtGPT2 extend to biocatalyst development and therapeutic protein design, where it accelerates the engineering cycle by reducing the need for extensive experimental screening [69].”
Gong, H., Wang, Y., Kong, Q., Li, X., Li, L., Wan, B., Zhao, Y., Zhang, J., Chen, G., Chen, J., et al. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.04.29.641945 · Cited by 1
“To prevent potential overfitting associated with models possessing a large number of parameters, such as ProGen [19, 43], we chose to use lightweight models following the previous work [21, 22], including Mamba [44] and Multi-head Latent Attention (MLA) [45], as shown in Fig.”
Leveraging State Space Models in Long Range Genomics
Popov, M., Kallala, A., Ramesh, A., Hennouni, N., Khaitan, S., Gentry, R., & Cohen, A.-S. (2025) · arXiv.org · doi:10.48550/arxiv.2504.06304 · Cited by 3
“…the effects of coding mutations on protein function (Lin et al., 2023), generate viable protein sequences conditioned on functional properties (Madani et al., 2023), and accurately predict protein structures (Lin et al., 2023), aided by benchmarks such as CASP (Kryshtafovych et al., 2021),…”
Leveraging Metamemory Agent for Enhanced Data-Free Code Generation in Large Language Models
Zhou, S., Wang, S., Ding, L., Zhan, Y.-B., Luo, Y., He, Z., Lin, F., & Tao, D.-P. (2025) · doi:10.48550/arxiv.2501.07892 · Cited by 3
Binary Discriminator Facilitates GPT-based Protein Design
Zeng, Z.-S., Xu, R., Guo, J., & Luo, X. (2024) · bioRxiv · doi:10.1101/2023.11.20.567789 · Cited by 3
“In the work of ProGen [12], evaluating the generated sequences was mainly done by log-likelihood scoring, which computes the per-residue conditional probabilities along the sequence, conceptually akin to perplexity, which measures the model's confidence in the generated sequence.”
MethodKBAlign: Efficient Self Adaptation on Specific Textual Knowledge Bases
Zeng, Z., Chen, Y., Yu, S., Wang, R., Yan, Y., Liu, Z., Wang, S., Han, X., Liu, Z., & Sun, M. (2025) · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP) · doi:10.18653/v1/2025.findings-emnlp.728
“For domains with plenty of data resources, researchers directly take domain materials in pre-training (Wang et al., 2023a; Madani et al., 2023).”
Shah, A., & Jayaratnam, S. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2411.05966 · Cited by 2
“17,19 ProteinMPNN uses a message passing neural network comprising of 3 encoder and 3 decoder layers and 128 hidden dimensions which predicts protein sequences in an autoregressive manner for a fixed backbone of interest.”
MethodRobust Neural Processes for Noisy Data
Shapira, C., & Rosenbaum, D. (2024) · arXiv.org · doi:10.48550/arxiv.2411.01670 · Cited by 3
“These models are used in different machine learning setups, including few shot learning [1, 2], meta learning [3, 4], language modeling [5] and generative modelling in various domains [6].”
MethodCai, F., Wei, Y., Kirchhofer, D., Chang, A. C., & Zhang, Y. (2024) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1012609 · Cited by 5
pLDDT-Predictor: High-speed Protein Screening Using Transformer and ESM2
Chae, J., Wang, Z., Gul, I., Ji, J., Chen, Z.-L., & Qin, P.-W. (2024) · arXiv.org · doi:10.48550/arxiv.2410.21283 · Cited by 3
“The rapid advancement of Large Language Models (LLMs)[3, 4] has further revolutionized the field of protein structure prediction and design[5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22].”
BackgroundUnlearning Virus Knowledge Toward Safe and Responsible Mutation Effect PredictionsPreprint
Li, M.-C., Zhou, B., Tan, Y., & Hong, L. (2024) · bioRxiv · doi:10.1101/2024.10.02.616274 · Cited by 3
“Decoder-only models, in comparison, are trained to optimize next-token prediction, which is frequently used for sequence design (Ferruz et al., 2022; Notin et al., 2022a; Madani et al., 2023).”
BackgroundMethodTowards deep learning sequence-structure co-generation for protein design
Wang, C.-T., Alamdari, S., Domingo-Enrich, C., Amini, A., & Yang, K. K. (2024) · doi:10.48550/arxiv.2410.01773 · Cited by 3
“…used to generate sequences with a specified function by conditioning on the desired structure [16], fine-tuning on a family of natural proteins with the desired function [61, 46, 62], conditioning on the desired reaction class [63, 64], and using gradients from a function prediction model [58, 65].”
MethodKalantar, M., Kalanther, I., Kumar, S., Buxton, E. K., & Raeeszadeh‐Sarmazdeh, M. (2024) · Computational and Structural Biotechnology Journal · doi:10.1016/j.csbj.2024.10.005 · Cited by 3
Fine-tuning of conditional Transformers for the generation of functionally characterized enzymesPreprint
Nicolini, M., Saitto, E., Franco, R. E. J., Cavalleri, E., Mesiti, M., Galeano, A., Malchiodi, D., Paccanaro, A., Robinson, P. N., Casiraghi, E., et al. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.08.10.607430 · Cited by 3
“In our experiments, we fine-tuned ProGen (Madani et al., 2023), a protein language model with 1 .”
BackgroundMethodResultVince, O., Gowers, G.-O. F., & McGibbon, S. (2024) · GEN Biotechnology · doi:10.1089/genbio.2024.0018 · Cited by 3
Huang, D., & Xie, J. (2024) · Molecules · doi:10.3390/molecules29122912 · Cited by 3
“The protein node features V p are computed using ESM-2b_650M to provide more biological knowledge [30].”
MethodShapeProt: Top-down Protein Design with 3D Protein Shape Generative ModelPreprint
Lee, Y., & Kim, J. (2023) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2023.12.03.567710 · Cited by 2
“Through distinguished pre-training like masked language modeling (MLM) [20] or causal language modeling (CLM) [21] on the sequence database, several protein language models, such as ESM [22] and ProGen [23], have been developed.”
MethodGeNLP: a web tool for NLP-based exploration and prediction of microbial gene function
Miller, D., Arias, O., & Burstein, D. (2024) · Bioinform. · doi:10.1093/bioinformatics/btae034 · Cited by 3
“Language models have emerged as a valuable tool for various biological data analysis tasks, including the analysis of genomic and protein sequences (Bepler and Berger 2021, Lin et al. 2023, Madani et al. 2023).”
Enhancing luciferase activity and stability through generative modeling of natural enzyme sequences
Xie, W.-J., Liu, D., Wang, X.-Y., Zhang, A., Wei, Q., Nandi, A., Dong, S.-W., & Warshel, A. (2023) · Proceedings of the National Academy of Sciences of the United States of America · doi:10.1073/pnas.2312848120 · Cited by 3
“Generative models, trained on natural homologous sequences, have shown potential in generating functional sequences akin to those found in nature (15–18).”
Diverse Preference Optimization
Lanchantin, J., Chen, A., Dhuliawala, S., Yu, P., Weston, J., Sukhbaatar, S., & Kulikov, I. (2025) · arXiv · doi:10.48550/arxiv.2501.18101 · Cited by 2
“Tasks such as creative writing, idea generation, and biological sequence design, explicitly require and benefit from diverse generations (Marco et al., 2024; Tachibana et al., 2025; Madani et al., 2023).”
Long-Term Energy Management for Large Model Training
Ma, J., Zhang, X., Sun, J., & Wang, Y. (2026) · IEEE Transactions on Industry Applications · doi:10.1109/tia.2026.3654489 · Cited by 2
“L ARGE models have shown strong performance in areas such as natural language processing [1] and protein structure prediction [2].”
A survey of LLMs in drug discovery and precision medicine
Hays, H., & Richardson, W. J. (2026) · Frontiers in Drug Discovery · doi:10.3389/fddsv.2026.1854899 · Cited by 2
FrustrAI-Seq: Scaling Local Energetic Frustration to the Protein Sequence SpacePreprint
Leusch, J.-P., Poley-Gil, M., Fernández-Martín, M., Simonetti, F. L., Schlensok, J., Bordin, N., Rost, B., Parra, R. G., & Heinzinger, M. (2026) · bioRxiv · doi:10.64898/2026.02.03.703498 · Cited by 2
“While ALM was shown to excel at generating functional proteins (21), MLM models have become the de facto standard solution for protein property prediction.”
Mechanist: AI as a Scientific Instrument for Discovering the Mechanisms of Intelligence
Wang, M.-R., Fang, J.-F., Qiao, S.-F., Xu, Z.-Q., Xu, H.-M., Wang, H.-X., Deng, S.-M., Yang, L.-Y., Cui, Z.-X., Xu, X., et al. (2026) · arXiv.org · doi:10.48550/arxiv.2608.12036 · Cited by 2
“Scientific foundation models are increasingly used not only to predict biological properties [37, 38], but also to generate molecules and genomic sequences with desired functions [39, 40, 41, 42].”
A Blind Spot in Alignment: Quantifying Biosecurity Risks in Large Language ModelsPreprint
Quan, S., Hao, T., Fang, S., Geng, H., Zhou, J., Chen, B., Wang, K., Hong, D., Dai, J., Yang, Y., et al. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2608.02684 · Cited by 1
“However, these mechanisms operate in natural-language space and cannot determine whether a compliant output—such as an amino acid sequence— encodes a biological threat (Madani et al., 2023; Nguyen et al., 2024; Watson et al., 2023).”
Generative artificial intelligence for enzyme design and biocatalysis
Middendorf, L., & Ferruz, N. (2026) · Current Opinion in Chemical Biology · doi:10.1016/j.cbpa.2026.102728 · Cited by 3
Microbial Enzyme Production: Critical Bottlenecks and Integrated Engineering Solutions
Maji, P., Sengupta, A., Ghosh, P., & Shukla, P. (2026) · Journal of Basic Microbiology · doi:10.1002/jobm.70180 · Cited by 2
ProtGPT3: an Open-source family of Promptable and Aligned Protein Language ModelsPreprint
Garibbo, M., Boxó, G., Stocco, F., Illanes-Vicioso, R., Middendorf, L., & Ferruz, N. (2026) · bioRxiv · doi:10.64898/2026.06.04.730041 · Cited by 2
“All our models were trained to perform causal language modeling in both directions [e.g., 32, 29, 4], enabling protein generation from either the N-to-C or C-to-N terminal with the help of two special directional tokens encoding the direction at the start of each protein (Appendix (A.2)).”
Antimicrobial Peptides Against Antimicrobial-Resistant Bacteria: Focus on Machine Learning
Tahmasebi, H., & Arabestani, M. R. (2026) · Infection and Drug Resistance · doi:10.2147/idr.s602699 · Cited by 2
“183 ProtBERT, a BERT-based model pre-trained on over 2 billion protein sequences, generates 1024-dimensional contextualised embeddings for each amino acid position.”
AMix-2: Establishing Protein as a Native Modality in Large Language ModelsPreprint
Qiu, K., Wu, Y., Wang, L., Ouyang, Y., Yu, J., Zhou, Z., Lv, C., Xue, D., Song, Y., Zhang, X., et al. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2605.30963 · Cited by 3
Detecting and quantifying overparametrization in RNA language models with REDIALPreprint
Teng, D., Qiu, Y.-R., Sakthivel, G., Aranganathan, A., Herron, L., & Tiwary, P. (2026) · bioRxiv · doi:10.64898/2026.05.11.724344 · Cited by 2
“In terms of architectures, debate continues regarding the relative merits of masked language models (MLMs) versus decoder-only causal architectures 56,57 , both paradigms have been extensively explored in the context of biomolecular sequences.”
Explicit representation of germline and non-germline residues improves antibody language modelingPreprint
Kim, J., Blalock, N., Kulkarni, A., Nakamura, K., & Romero, P. A. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.05.06.723387 · Cited by 1
“While general protein language models (pLMs) have established a foundation for biological sequence modeling (Lin et al., 2023; Madani et al., 2023), the unique generative constraints of immunoglobulins necessitate specialized antibody language models (ALMs) (Olsen et al., 2022; Shuai et al., 2023).”
Calvanese, F., Lombardi, G., Weigt, M., & FERNANDEZ-DE-COSSIO-DIAZ, J. (2026) · bioRxiv · doi:10.64898/2026.05.01.722156 · Cited by 2
“Protein Language Models (pLMs) trained on millions of natural proteins have demonstrated a remarkable ability to capture complex evolutionary rules governing protein struc-38 ture and function (Madani et al., 2023; Lin et al., 2023).”
InfluentialGenerative Protein Design: From Deep Learning Algorithms to Translational Applications
Luo, S., & Zhou, B. (2026) · International Journal of Molecular Sciences · doi:10.3390/ijms27093917 · Cited by 1
“Therefore, it can be seen that the sequence semantics can play a role in the generative task, and can make up for the lack of physical constraints, especially in the absence of homologous templates [18–20].”
Sequence Display Enables Large-Scale Sequence–Activity Datasets for Rapid Protein Evolution
Cheng, L.-Q., Zheng, X., Jiang, S., Hu, Y., Liu, Y.-J., Yang, K.-Q., Rui, J.-Y., Ding, H.-X., Zhang, M., Yuan, T., et al. (2026) · Nature Biotechnology · doi:10.1038/s41587-026-03087-3 · Cited by 2
Han, H., Zhang, Q., Chen, Y., Wang, Z.-X., Su, X.-Y., Zhu, Y., Wang, X., & Feng, Z.-X. (2026) · International Journal of Biological Macromolecules · doi:10.1016/j.ijbiomac.2026.152018 · Cited by 2
Inference-Time Toxicity Mitigation in Protein Language Models
Burda, M. F., Aranguri, S., Moreno, I. A., & Ferrante, E. (2026) · arXiv.org · doi:10.48550/arxiv.2603.04045 · Cited by 2
“Protein language models (PLMs) like ESM-2 (Lin et al., 2023a) and ProGen (Madani et al., 2023; Nijkamp et al., 2022; Bhatnagar et al., 2025) have demonstrated the capacity to generate novel proteins with predicted functionality, whilst genomic models like Evo2 have enabled the in silico design of synthetic bacteriopha…”
InfluentialRen, C.-Y., He, D., & Huang, J. (2026) · Briefings Bioinform. · doi:10.1093/bib/bbag095 · Cited by 2
“Evolutionary scale modeling (ESM) [19] leverages transformer architectures to capture evolutionary patterns in protein sequences [20] developed an LLM capable of generating amino acid sequences based on specific keywords for various protein families.”
AI-driven computational methods and benchmarking for T-cell antigen identification
Deng, Y., Que, J., Xue, G., Cai, Y., Yang, W., Wang, Y., Yi, H., Wang, Z., Lin, Y., Zhou, W., et al. (2026) · Briefings in Bioinformatics · doi:10.1093/bib/bbag123 · Cited by 1
Rational design of synthetic proteins using a genome-scale CRISPR screenPreprint
Burrell, W. H., Mueller, S. J., Daniloski, Z., Doyle, P. D., Rovsing, A. B., James, C., Drabkin, M., Chou, C.-Y., So, H. Y. A., Katgara, L., et al. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.02.19.706875 · Cited by 2
“Others have paired protein sequences with natural language descriptions and, using large language models, have generated novel sequences with specific functional properties 7 , including programmable CRISPR nucleases similar to the widely-used S. pyogenes Cas9 8 .”
Generative AI for Enzyme Design and BiocatalysisPreprint
Middendorf, L., & Ferruz, N. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2602.03779 · Cited by 2
“While masked-language models are typically used to score and prioritize a relatively small number of substitutions in a given scaffold, autoregressive sequence-generating models can propose entirely new sequences and thereby explore a broader region of sequence space[28, 29, 30, 31, 32, 33, 34].”
Large-scale multi-omic biosequence transformers for modeling protein–nucleic acid interactions
Chen, S. F., Steele, R. J., Hocky, G. M., Lemeneh, B., Lad, S. P., & Oermann, E. K. (2026) · PLoS ONE · doi:10.1371/journal.pone.0341501 · Cited by 3
Zhang, C.-W., Yang, L., Qin, Y.-J., Li, D., Dong, S., & Yang, M. (2026) · bioRxiv · doi:10.64898/2026.02.01.703093 · Cited by 2
Identifying microbial protease allergens through protein language model-guided homology
Thurimella, K., Wu, E., Li, C., Graham, D. B., Owens, R. M., Plichta, D. R., Sokol, C. L., Xavier, R. J., & Bacallado, S. (2026) · Cell Systems · doi:10.1016/j.cels.2025.101510 · Cited by 1
Ye, A., Wang, B.-Y., Cheng, Z., Dong, C., & Guo, F.-B. (2026) · Bioresource Technology · doi:10.1016/j.biortech.2026.134309 · Cited by 2
Yang, L., He, Y. J., Cao, F., Qin, Y., Wang, Y., Zhang, H., Han, W., & Yang, M. (2026) · Nucleic Acids Research · doi:10.1093/nar/gkaf1529 · Cited by 2
“The RF parameters were as follows: {‘n_estimators’: [25, 50, 100, 200, 400], “min_samples_split”: [2, 4, 8], “min_samples_leaf”: [1, 2, 4, 8], “max_features”: [“sqrt”, “log2”, None]}.”
Wang, J., Geng, A.-Y., Yan, Z., Lin, X., Li, Y.-H., Xu, J.-L., Meng, Y.-J., Zhang, Z.-L., Wei, L.-Y., Zou, Q., et al. (2025) · International Journal of Biological Macromolecules · doi:10.1016/j.ijbiomac.2025.149267 · Cited by 2
ApexGen: Simultaneous design of peptide binder sequence and structure for target proteins.Preprint
Xia, X.-Q., & de la Fuente-Núñez, C. (2025) · arXiv.org · doi:10.48550/arxiv.2511.14663 · Cited by 2
Hou, C., Liu, D., & Shen, Y.-F. (2025) · bioRxiv · doi:10.1101/2025.11.04.686576 · Cited by 2
g-DPO: Scalable Preference Optimization for Protein Language Models
Ferragu, C., Ziegler, J. D., Deutschmann, N., Lindoulsi, A., Bixby, E., & Cradle ML Team (2025) · arXiv.org · doi:10.48550/arxiv.2510.19474 · Cited by 2
“Protein language models (PLMs), trained with self-supervised learning on large sequence datasets, capture relevant structure and function signals [14, 17, 25, 33].”
Liao, D.-Q., Liu, C., Sun, X.-Z., Tang, D., Wang, H.-C., Youlten, S. E., Gopinath, S. K., Lee, H., Strayer, E., Giráldez, A., et al. (2025) · arXiv.org · doi:10.48550/arxiv.2510.24736 · Cited by 2
“De novo generative models excel at creating novel sequences, but fundamentally operate by generating from scratch rather than refining existing functional sequences (Prykhodko et al., 2019; M´endez-Lucio et al., 2020; Dauparas et al., 2022; Wu et al., 2021; Madani et al., 2023; Watson et al., 2023).”
Sequence-Based Generative AI-Guided Design of Versatile Tryptophan SynthasesPreprint
Lambert, T., Tavakoli, A., Dharuman, G., Yang, J., Bhethanabotla, V. C., Kaur, S., Hill, M., Ramanathan, A., Anandkumar, A., & Arnold, F. H. (2025) · bioRxiv · doi:10.1101/2025.08.30.673177 · Cited by 2
“9,13,14 19–22 Building on this promise, we used the GenSLM model and developed a filtering pipeline to prioritize candidates with the highest potential for experimental validation.”
Shi, H., Lin, Y., Liu, D., & Zou, Q. (2026) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1013397 · Cited by 2
Predicting Subcutaneous Antibody Bioavailability Using Ensemble Protein Language Models.
Cabreza, M., Hojegian, W., Wu, I.-E., & Lai, P.-K. (2025) · Molecular Pharmaceutics · doi:10.1021/acs.molpharmaceut.5c00523 · Cited by 2
Considering Metabolic Context in Enzyme Evolution and Design
Dokwal, D., Brown, P. M., Filipowska, K., & Reynolds, K. A. (2025) · Biochemistry · doi:10.1021/acs.biochem.5c00165 · Cited by 2
“34,133,134 Machine learning models trained on sequence have designed new enzymes, 34,133 fluorescent proteins, 135 and much more.”
Heuristic multi-site optimization for protein sequence design using Masked Protein Language Models
Wang, L., Wang, Y.-Z., Qiu, C., Xiao, L., Liu, X.-L., & Chen, J.-J. (2025) · bioRxiv · doi:10.1101/2025.07.31.668012 · Cited by 2
“The directed evolution (DE) framework is a cornerstone of protein sequence optimization through iterative cycles of random mutagenesis, high-throughput screening, and selective pressure to optimize sequences [3,4].”
An adaptive refinement neural particle method for granular flows
Pai, P.-H., Sarno, L., Tai, Y.-C., & Kan, H.-C. (2025) · The Physics of Fluids · doi:10.1063/5.0276235 · Cited by 2
Yang, Z., & Tamura, T. (2025) · IEEE Transactions on Computational Biology and Bioinformatics · doi:10.1109/tcbbio.2025.3590141 · Cited by 2
“Althoughadvanceddeeplearning models,suchasrecentlargelanguagemodels(LLMs)and otherfoundationmodelsspecializedformolecularorprotein sequences[38],[39],oftenachievesuperiorperformancein representationlearningtasks,theycomewithasignificantly largernumberofparameterscomparedtoclassicalsequential…”
Enhancing Safe and Controllable Protein Generation via Knowledge Preference Optimization
Wang, Y., Ding, K., Feng, K., Wang, Z., Qin, M., Li, X., Zhang, Q., & Chen, H. (2025) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2025.acl-long.616
“For example, in enzyme engineering, pretrained models predict mutations that enhance catalytic efficiency or substrate specificity, significantly accelerating the iterative design process [Madani et al., 2023; Zhou et al., 2024].”
mamp-ml: A deep learning approach to epitope immunogenicity in plantsPreprint
Stevens, D. M., Yang, D., Liang, T., Li, T., Vega, B., Coaker, G., & Krasileva, K. (2025) · bioRxiv · doi:10.1101/2025.07.11.664399 · Cited by 2
DivPro: diverse protein sequence design with direct structure recovery guidance
Zhou, X., Shen, G., Chen, Y.-C., Chen, G., & Heng, P. (2025) · Bioinformatics · doi:10.1093/bioinformatics/btaf258 · Cited by 2
“Recently, protein language models have demonstrated great success across various downstream tasks, including sequence design and structure prediction (Lin et al. 2023, Madani et al. 2023).”
Multi-Objective Structure-Based Drug Design Using Causal Discovery
Zhou, J., Zhao, D., Qian, H., Tu, S., & Xu, L. (2025) · IEEE Transactions on Computational Biology and Bioinformatics · doi:10.1109/tcbbio.2025.3572178
“Furthermore, the advancement of large language models (LLMs) has facilitated the development of specialized foundation models (FMs) for various biological entities, such as drug molecule [9], [43], [44], protein [45], [46], [47], RNA [48], [49], [50], [51], etc.”
Advances in Intelligent Design and Optimization Methods for Nucleic Acid Sequences
Yang, T., Han, M., Wen, X., & Zheng, Y. (2025) · Journal of Artificial Intelligence in Bioinformatics · doi:10.62762/jaib.2025.194547 · Cited by 3
Efficient Design of Affilin® Protein Binders for HER3
Díaz-Rovira, A. M., Lotze, J., Hoffmann, G., Pallara, C., Molina, A., Coburger, I., Gloser-Bräunig, M., Meysing, M., Zwarg, M., Díaz, L., et al. (2025) · International Journal of Molecular Sciences · doi:10.3390/ijms26104683 · Cited by 2
Sequence and taxonomic feature evaluation facilitated the discovery of alcohol oxidases
Han, Y., Ding, X., Tan, J., Sun, Y., Duan, Y., Liu, Z., Zheng, G., & Lu, D. (2025) · Synthetic and Systems Biotechnology · doi:10.1016/j.synbio.2025.04.014 · Cited by 1
“Recent years have witnessed the rapid development of data technology for enzyme discovery and design for unprecedented applications [1 – 3].”
Harnessing A Unified Multi-modal Sequence Modeling to unveil Protein-DNA InterdependencyPreprint
Li, M., Ren, Y., Ye, P., Cheng, J., Ma, X., Cai, Y., Ouyang, W., Zhong, B., Wu, B., Dong, N., et al. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.02.26.640480 · Cited by 1
“As shown in Figure 2B, when ProDMM utilizes protein sequences as input, it demonstrates superior performance compared to other protein sequence models, such as ESM2 and progen [5, 14].”
A Feature Engineering Method for Whole-Genome DNA Sequence with Nucleotide Resolution
Wang, T., Cui, Y.-P., Sun, T., Li, H., Wang, C., Hou, Y., Wang, M., Chen, L., & Wu, J.-M. (2025) · International Journal of Molecular Sciences · doi:10.3390/ijms26052281 · Cited by 2
Steering Protein Family Design through Profile Bayesian Flow
Gong, J., Pei, Y., Long, S., Song, Y., Zhang, Z., Huang, W., Cao, Z., Zhang, S.-Y., Zhou, H., & Ma, W.-Y. (2025) · International Conference on Learning Representations · doi:10.48550/arxiv.2502.07671 · Cited by 2
“Hence, there is also rising interest in transferring the techniques from language modeling to protein modeling (Truong Jr & Bepler, 2023; Madani et al., 2023; 2020; Nijkamp et al., 2023; Jumper et al., 2021).”
Transformation of Knowledge-Centered Pedagogy with ChatGPT and AI in Educational Practices
Khalid, A. (2025) · European Journal of Applied Science Engineering and Technology · doi:10.59324/ejaset.2025.3(1).05 · Cited by 3
Interpretable adenylation domain specificity prediction using protein language modelsPreprint
Adduri, A. K., McNutt, A. T., Ellington, C. N., Suraparaju, K., Fang, N., Yan, D., Krummenacher, B., Li, S.-T., Bodden, C., Xing, E., et al. (2025) · bioRxiv · doi:10.1101/2025.01.13.632878 · Cited by 2
“As such, protein language models have been used to guide protein design and generation (41–43) and predict drug-target interactions (44–46).”
Imam, I. A., Bailey, S., Wang, D., Zeng, S., Xu, D. H., & Shao, Q. (2024) · Langmuir · doi:10.1021/acs.langmuir.4c04140 · Cited by 5
Large language modeling and deep learning shed light on RNA structure prediction
(2024) · Nature Methods · doi:10.1038/s41592-024-02488-z · Cited by 2
Incremental Inverse Design of Desired Soybean Phenotypes
Zavorskas, J., Edwards, H., Marten, M. R., Harris, S. D., & Srivastava, R. (2024) · ACS Omega · doi:10.1021/acsomega.4c01704 · Cited by 2
Tomilova, Y. E., Russkikh, N., Yi, I. M., Shaburova, E., Tomilov, V. N., Pyrinova, G. B., Brezhneva, S. O., Tikhonyuk, O. S., Gololobova, N. S., Popichenko, D. V., et al. (2024) · bioRxiv · doi:10.1101/2024.07.24.604875 · Cited by 2
Artificial intelligence-driven health research innovations: Protein sciences
Liu, F., Zhang, G., Liu, Z., Li, C., & Huang, X. (2024) · Medicine Plus · doi:10.1016/j.medp.2024.100043 · Cited by 2
Large-Scale Multi-omic Biosequence Transformers for Modeling Protein-Nucleic Acid Interactions
Chen, S. F., Steele, R. J., Hocky, G. M., Lemeneh, B., Lad, S. P., & Oermann, E. (2024) · arXiv.org · doi:10.48550/arxiv.2408.16245 · Cited by 2
Complex-based Ligand-Binding Proteins Redesign by Equivariant Diffusion-based Generative ModelsPreprint
Nguyen, V. T. D., Nguyen, N. D., & Hy, T. (2024) · bioRxiv · doi:10.1101/2024.04.17.589997 · Cited by 2
“…of PLMs is that sequences of amino acids can be conceptualized similarly to sentences composed of words, enabling the application of advanced text processing techniques to predict the structural, functional, and interactional properties of proteins based solely on their amino acid sequences.”
BackgroundIntelligent Design of Antithrombotic Peptide Targeting Collagen
Song, C., & Zhang, L. (2024) · Langmuir · doi:10.1021/acs.langmuir.4c00543 · Cited by 3
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
