The DOI matches its Crossref record; 130 of the 133 references it cites were found in a registry or an index; none of the 53 sources its claims cite is checked for support yet.
Learning the protein language: Evolution, structure, and function
Open accessAbstract
Language models have recently emerged as a powerful machine-learning approach for distilling information from massive protein sequence databases. From readily available sequence data alone, these models discover evolutionary, structural, and functional organization across protein space. Using language models, we can encode amino-acid sequences into distributed vector representations that capture their structural and functional properties, as well as evaluate the evolutionary fitness of sequence variants. We discuss recent advances in protein language modeling and their applications to downstream protein property prediction problems. We then consider how these models can be enriched with prior biological knowledge and introduce an approach for encoding protein structural knowledge into the learned representations. The knowledge distilled by these models allows us to improve downstream function prediction through transfer learning. Deep protein language models are revolutionizing protein biology. They suggest new ways to approach protein and therapeutic design. However, further developments are needed to encode strong biological priors into protein language models and to increase their accessibility to the broader community.
Overview
AI snapshot
Written from the paper’s full text.
TL;DR
Protein language models learn evolutionary, structural, and functional organization from large sequence databases, and adding structure supervision improves downstream prediction through transfer learning. The authors present a multi-task LSTM trained on 76 million sequences plus 28,000 structures and show gains in structural similarity, transmembrane prediction, and sequence-to-phenotype prediction.
Key findings
- A multi-task LSTM (MT-LSTM) trained with masked language modeling, residue-residue contact prediction, and SCOP structural similarity prediction learned embeddings that “dramatically outperform” sequence comparison methods and even outperform TMalign for structural similarity; structure supervision was critical for fine-grained organization, and MT-LSTM outperformed the prior SSA-LSTM.
- For transmembrane region prediction, BiLSTM+CRF models using MT-LSTM embeddings significantly outperformed DLM-LSTM embeddings (paired t-test, p=0.044) and also outperformed existing predictors and SSA-LSTM.
- For sequence-to-phenotype prediction on deep mutational scanning datasets, MT-LSTM embeddings with Gaussian process regression achieved an average Spearman correlation of 0.65, significantly outperforming ECNet’s 0.60 (paired t-test, p=0.006).
- The paper notes that language models predicted SARS-CoV-2 spike mutations with high antibody-escape potential, but as of publication viral escape had not yet been experimentally verified.
Methods
- The authors trained a three-layered bidirectional LSTM with skip connections and 1024 hidden units in each direction per layer. The MT-LSTM combined three losses: masked language modeling (10% of amino acids randomly masked), residue-residue contact prediction (contacts defined as residues within 8 Å), and SCOP structural similarity prediction via ordinal regression; MT-LSTM loss weights were λmasked=0.5, λcontact=0.9, λsimilarity=0.1, while DLM-LSTM used only λmasked=1.
- Training data were 76,215,872 UniRef90 sequences (July 2018) and 28,010 SCOPe ASTRAL 2.06 protein sequences with structures, split into 22,408 training and 5,602 testing sequences. Models were trained for 1,000,000 parameter updates with Adam at learning rate 0.0001 on a single NVIDIA V100 GPU; training took roughly 13 days for DLM-LSTM and 51 days for MT-LSTM.
- Evaluations included structural similarity on 100,000 randomly sampled SCOPe ASTRAL test pairs, transmembrane prediction on the TOPCONS2 dataset of 5,154 proteins using 10-fold cross validation, and sequence-to-phenotype prediction on deep mutational scanning datasets using 5-fold cross validation and Gaussian process regression with an RBF kernel.
Limitations
- The paper states that further developments are needed to encode strong biological priors into protein language models and to increase their accessibility to the broader community; it also states that current models make heavily simplified phylogenetic assumptions, treating sequences as independent draws from a star phylogeny.
- The paper states practical limitations including high compute costs, murky design guidelines, ad hoc model development, quadratic scaling of transformers with sequence length (requiring long sequences to be excluded or truncated), and poor learning of long-range dependencies by RNNs.
- Evident from the text: the new results come from a single MT-LSTM architecture and specific benchmarks, and the SARS-CoV-2 escape predictions had not been experimentally verified as of publication.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
0 of 53 cited sources checked · 0 supported · 0 flagged
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 find that increasing model size improves transfer learning performance.”
§ Learning the Protein Language: Evolution, Structure and Function > Transfer learning improves downstream applications
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesA 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
“However, this organization is improved when we include structure supervision in the language model training (Figure 4b).”
§ Learning the Protein Language: Evolution, Structure and Function > Multi-task language models capture the semantic organization of proteins
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesA 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
“Furthermore, representations learned by our MT-LSTM model significantly outperform (paired t-test, p=0.044) the embeddings learned by our DLM-LSTM model on this application (Figure 5b).”
§ Learning the Protein Language: Evolution, Structure and Function > Transfer learning improves downstream applications
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIn the paper
“Furthermore, the multi-task learning approach outperforms a two-step learning approach presented previously (SSA-LSTM)27.”
§ Learning the Protein Language: Evolution, Structure and Function > Multi-task language models capture the semantic organization of proteins
What the sources say
- Not checked yetcites [27] Bepler T & Berger B Learning protein sequence embeddings using information from structure. in International Conference on Learning Representations (2019). PMC12947628 41768824 source
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesA 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
“Because we embed sequences into a semantic representation space, we can find structurally related proteins even though their sequences are not closely related (Figure 4c, Supplemental Table 1).”
§ Learning the Protein Language: Evolution, Structure and Function > Multi-task language models capture the semantic organization of proteins
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesA 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 observe that even though the DLM-LSTM model was trained using only sequence information, protein sequences still organize roughly by structure in embedding space.”
§ Learning the Protein Language: Evolution, Structure and Function > Multi-task language models capture the semantic organization of proteins
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesA 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
“Because TMscore is not symmetric, we calculate TMscore for both comparison directions and average them together for each protein pair. We found this outperformed other methods of combining the two scores.”
§ Learning the Protein Language: Evolution, Structure and Function > Methods > Protein structural similarity prediction evaluation
Strength:plain· not yet calibratedCausal:direct· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes? Load-bearing: unclear✓ Based: passesIn the paper
“The key insight was that the sequence representations (vector embeddings) learned by the language model were powerful features for solving other prediction problems.”
§ Learning the Protein Language: Evolution, Structure and Function > Transfer learning improves downstream applications
What the sources say
- Not checked yetcites [27] Bepler T & Berger B Learning protein sequence embeddings using information from structure. in International Conference on Learning Representations (2019). PMC12947628 41768824 source
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIn the paper
“Interestingly, because so many sequences are available, these models continue to benefit from increased size 32.”
§ Learning the Protein Language: Evolution, Structure and Function > Protein language models distill information from massive protein sequence databases
What the sources say
- Not checked yetcites [32] Rives A Biological structure and function emerge from scaling unsupervised learning to 250 million protein sequences. bioRxiv (2019). PMC8053943 33876751 source
Strength:plain· not yet calibratedCausal:direct· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes? Load-bearing: unclear✓ Based: passesIn the paper
“Learned protein representations improve predictive ability for downstream prediction problems through transfer learning (Figure 5a).”
§ Learning the Protein Language: Evolution, Structure and Function > Transfer learning improves downstream applications
What the sources say
- Not checked yetcites [27] Bepler T & Berger B Learning protein sequence embeddings using information from structure. in International Conference on Learning Representations (2019). PMC12947628 41768824 source
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passes- Rests on an assumption: Many problems in biology are small data problems; quantitative protein characterization assays are rarely high throughput and methods are needed that can generalize given only 10s to 100s of experimental measurements.
- Rests on an assumption: We are often interested in extrapolating from data collected over a small region of protein sequence space to other sequences, often with little to no homology.
- It may state two separate claims as one. 13
- The quote only partly states this claim. 1
- Probably not a substantial claim. 8
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 find that our learned semantic embeddings dramatically outperform the sequence comparison methods and even outperform structure comparison with TMalign when predicting structural similarity.”
§ Learning the Protein Language: Evolution, Structure and Function > Multi-task language models capture the semantic organization of proteins
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt 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 knowledge distilled by these models allows us to improve downstream function prediction through transfer learning.”
- Also stated in: § Multi-task language models capture the semantic organization of proteins · boosted
“Here, we will demonstrate that training protein language models with self-supervision on a large amount of natural sequence data and with structure supervision on a smaller set of sequence, structure pairs enriches the learned representations and translates into improvements in downstream prediction problems (Figure 3).”
- Also stated in: § Transfer learning improves downstream applications · plain
“Learned protein representations improve predictive ability for downstream prediction problems through transfer learning (Figure 5a).”
§ Learning the Protein Language: Evolution, Structure and Function > Abstract
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one.
- Rests on an assumption: Many problems in biology are small data problems; quantitative protein characterization assays are rarely high throughput and methods are needed that can generalize given only 10s to 100s of experimental measurements.
- Rests on an assumption: We are often interested in extrapolating from data collected over a small region of protein sequence space to other sequences, often with little to no homology.
- Also stated in: § Multi-task language models capture the semantic organization of proteins · boosted
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 find that we can predict the phenotypes of unobserved sequence variants across datasets better than existing methods (Figure 5c).”
§ Learning the Protein Language: Evolution, Structure and Function > Transfer learning improves downstream applications
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one. The quote only partly states this claim.
- Rests on an assumption: Many problems in biology are small data problems; quantitative protein characterization assays are rarely high throughput and methods are needed that can generalize given only 10s to 100s of experimental measurements.
- Rests on an assumption: We are often interested in extrapolating from data collected over a small region of protein sequence space to other sequences, often with little to no homology.
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 find that the BiLSTM+CRFs with our new embeddings (DLM-LSTM and MT-LSTM) outperform existing transmembrane predictors and a BiLSTM+CRF using our previous smaller embedding model (SSA-LSTM).”
§ Learning the Protein Language: Evolution, Structure and Function > Transfer learning improves downstream applications
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt 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
“Interestingly, we observe that the structural supervision component is critical for learning well organized embeddings at a fine-grained level, because the DLM-LSTM representations alone do not perform well at this task (Supplemental Table 1).”
§ Learning the Protein Language: Evolution, Structure and Function > Multi-task language models capture the semantic organization of proteins
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one.
In the paper
“This parallels the general trend in natural language processing, where the number of parameters, rather than specific architectural choices, is the best indicator of model performance 52.”
§ Learning the Protein Language: Evolution, Structure and Function > Protein language models distill information from massive protein sequence databases
What the sources say
- Not checked yetcites [52] Kaplan J Scaling Laws for Neural Language Models. arXiv [cs.LG] (2020).
Strength:plain· not yet calibratedCausal:correlational· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✗ Load-bearing: fails✓ Based: passesProbably not a substantial claim. The paper’s conclusions may not depend on it.
None of the sources cited for this claim have been checked yet.
In the paper
“Although structure is more informative of function than sequence 57,58 and structure is encoded by sequence, predicting structure remains hard, particularly due to the relative paucity of structural relative to sequence data.”
§ Learning the Protein Language: Evolution, Structure and Function > Supervision encodes biological meaning
What the sources say
- Not checked yetcites [57] Zhang C & Kim S-H Overview of structural genomics: from structure to function. Curr. Opin. Chem. Biol 7, 28– 32 (2003). 12547423 source
- Not checked yetcites [58] Shin DH Structure-based inference of molecular functions of proteins of unknown function from Berkeley Structural Genomics Center. J. Struct. Funct. Genomics 8, 99– 105 (2007). 17764033 source
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one.
None of the sources cited for this claim have been checked yet.
In the paper
“They also learn summary representations, powerful features that can be used to better capture sequence relationships and link sequence to function via transfer learning 27,28,32-35.”
§ Learning the Protein Language: Evolution, Structure and Function > Introduction
What the sources say
- Not checked yetcites [27] Bepler T & Berger B Learning protein sequence embeddings using information from structure. in International Conference on Learning Representations (2019). PMC12947628 41768824 source
- Not checked yetcites [28] Alley EC, Khimulya G, Biswas S, AlQuraishi M & Church GM Unified rational protein engineering with sequence-based deep representation learning. Nat. Methods 16, 1315– 1322 (2019). 31636460 PMC7067682 source
- Not checked yetcites [32] Rives A Biological structure and function emerge from scaling unsupervised learning to 250 million protein sequences. bioRxiv (2019). PMC8053943 33876751 source
- Not checked yetcites [33] Rao R Evaluating Protein Transfer Learning with TAPE. Adv. Neural Inf. Process. Syst 32, 9689– 9701 (2019). 33390682 PMC7774645 source
- Not matchedcites [34] Luo Y Evolutionary Context-Integrated Deep Sequence Modeling for Protein Engineering. in Research in Computational Molecular Biology 261– 263 (Springer International Publishing, 2020).
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not checked yetcites [35] Hie B, Bryson BD & Berger B Leveraging Uncertainty in Machine Learning Accelerates Biological Discovery and Design. Cell Syst 11, 461– 477.e9 (2020). 33065027 source
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one.
None of the sources cited for this claim have been checked yet.
In the paper
“Although it is generally believed that proteins with similar sequences form similar structures, there are also interesting examples of highly similar protein sequences having radically different structures and functions 60,61 and of sequences that can form multiple folds 62.”
§ Learning the Protein Language: Evolution, Structure and Function > Supervision encodes biological meaning
What the sources say
- Not checked yetcites [60] Kosloff M & Kolodny R Sequence-similar, structure-dissimilar protein pairs in the PDB. Proteins 71, 891– 902 (2008). 18004789 PMC2673347 source
- Not checked yetcites [61] Wei KY Computational design of closely related proteins that adopt two well-defined but structurally divergent folds. Proc. Natl. Acad. Sci. U. S. A 117, 7208– 7215 (2020). 32188784 PMC7132107 source
- Not checked yetcites [62] James LC & Tawfik DS Conformational diversity and protein evolution--a 60-year-old hypothesis revisited. Trends Biochem. Sci 28, 361– 368 (2003). 12878003 source
Strength:plain· not yet calibratedCausal:correlational· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes? Load-bearing: unclear✓ Based: passesIt 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
“For example, unrelated protein sequences can form the same structural fold and, therefore, are semantically similar. However, we cannot deduce this relationship from sequences alone. Supervision is required to learn that these sequences belong to the same semantic category.”
§ Learning the Protein Language: Evolution, Structure and Function > Supervision encodes biological meaning
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt 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 phenomena suggest that there are aspects of protein biology that may not be discoverable by statistical sequence models alone. Supervision that represents known protein structure, function, and other prior knowledge may be necessary to encode distant sequence relationships into learned embeddings.”
§ Learning the Protein Language: Evolution, Structure and Function > Supervision encodes biological meaning
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one.
- Rests on an assumption: Proteins with similar structures and functions still exist below 30% sequence identity, but they can no longer be detected from sequence similarity alone and it is unclear whether their functions are conserved.
- Rests on an assumption: Evolutionary innovation requires that protein function can change with only a few mutations.
None of the sources cited for this claim have been checked yet.
In the paper
“Language models learn the probability of a sequence occurring and this can be directly applied to predict the fitness of sequence mutations 29-31 .”
§ Learning the Protein Language: Evolution, Structure and Function > Introduction
What the sources say
- Not checked yetcites [29] Riesselman AJ, Ingraham JB & Marks DS Deep generative models of genetic variation capture the effects of mutations. Nat. Methods 15, 816– 822 (2018). 30250057 PMC6693876 source
- Not checked yetcites [30] Hie B, Zhong E, Bryson B & Berger B Learning mutational semantics. Adv. Neural Inf. Process. Syst 33, (2020).
- Not checked yetcites [31] Hie B, Zhong ED, Berger B & Bryson B Learning the language of viral evolution and escape. Science 371, 284– 288 (2021). 33446556 source
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt 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
“Language models capture complex relationships between residues in protein sequences by condensing information from enormous protein sequence databases.”
§ Learning the Protein Language: Evolution, Structure and Function > Conclusions and Perspectives: Strong biological priors are key to improving protein language models
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✗ Based: failsProbably not a substantial claim. The paper may give no reason why it holds.
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
“Fine-tuning pre-trained models can help with this problem but does not fully resolve it.”
§ Learning the Protein Language: Evolution, Structure and Function > Conclusions and Perspectives: Strong biological priors are key to improving protein language models
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesProbably not a substantial claim. It may state two separate claims as one.
- Rests on an assumption: For individual protein families or application-specific protein datasets, brute force hyperparameter search is the gold standard for selecting model architectures and number of parameters.
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
“High compute costs and murky design guidelines mean that developing new models is often an expensive, time consuming, and ad hoc process.”
§ Learning the Protein Language: Evolution, Structure and Function > Conclusions and Perspectives: Strong biological priors are key to improving protein language models
Strength:hedged· not yet calibratedCausal:direct· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✗ Load-bearing: fails✓ Based: passesProbably not a substantial claim. The paper’s conclusions may not depend on it.
None of the sources cited for this claim could be matched to a paper in Crossref, so there is nothing to check it against.
In the paper
“To compare methods, we follow Luo et al and perform 5-fold cross validation on each deep mutational scanning dataset 34”
§ Learning the Protein Language: Evolution, Structure and Function > Methods > Sequence-to-phenotype prediction and evaluation
What the sources say
- Not matchedcites Luo et al
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [34] Luo Y Evolutionary Context-Integrated Deep Sequence Modeling for Protein Engineering. in Research in Computational Molecular Biology 261– 263 (Springer International Publishing, 2020).
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
Load-bearingStrength:plain· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesProbably not a substantial claim.
- Not matched
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 compare methods by 10-fold cross validation.”
§ Learning the Protein Language: Evolution, Structure and Function > Methods > Transmembrane region prediction training and evaluation
Load-bearingStrength:plain· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesProbably not a substantial claim.
None of the sources cited for this claim have been checked yet.
In the paper
“New linear complexity attention mechanisms may help to alleviate this limitation 95,96.”
§ Learning the Protein Language: Evolution, Structure and Function > Conclusions and Perspectives: Strong biological priors are key to improving protein language models
What the sources say
- Not matchedcites [95] Choromanski KM Rethinking Attention with Performers. in International Conference on Learning Representations (2020).
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not checked yetcites [96] Wang S, Li BZ, Khabsa M, Fang H & Ma H Linformer: Self-Attention with Linear Complexity. (2020). source
Strength:hedged· not yet calibratedCausal:conditional· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✗ Load-bearing: fails✓ Based: passesProbably not a substantial claim. The paper’s conclusions may not depend on it.
- Rests on an assumption: Transformers scale quadratically with sequence length.
- Not matched
In the paper
“Recently, Luo et al. presented a method for combining language model-based representations with local evolutionary context-based representations (ECNet) and demonstrated that these representations were powerful for sequence-to-phenotype mapping on a panel of deep mutational scanning datasets 34.”
§ Learning the Protein Language: Evolution, Structure and Function > Transfer learning improves downstream applications
What the sources say
- Not matchedcites [34] Luo Y Evolutionary Context-Integrated Deep Sequence Modeling for Protein Engineering. in Research in Computational Molecular Biology 261– 263 (Springer International Publishing, 2020).
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
Strength:boosted· not yet calibratedCausal:correlational· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✗ Load-bearing: fails✓ Based: passesProbably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
- Not matched
The paper relies on this without support.
Many problems in biology are small data problems; quantitative protein characterization assays are rarely high throughput and methods are needed that can generalize given only 10s to 100s of experimental measurements.
“A key challenge in biology is that many problems are small data problems. Quantitative protein characterization assays are rarely high throughput and methods are needed that can generalize given only 10s to 100s of experimental measurements.”
The paper relies on this without support.
We are often interested in extrapolating from data collected over a small region of protein sequence space to other sequences, often with little to no homology.
“Furthermore, we are often interested in extrapolating from data collected over a small region of protein sequence space to other sequences, often with little to no homology.”
The paper relies on this without support.
Proteins with similar structures and functions still exist below 30% sequence identity, but they can no longer be detected from sequence similarity alone and it is unclear whether their functions are conserved.
“Often called the “twilight zone” of protein sequence homology, proteins with similar structures and functions still exist below this level, but they can no longer be detected from sequence similarity alone and it is unclear whether their functions are conserved.”
The paper relies on this without support.
Evolutionary innovation requires that protein function can change with only a few mutations.
“Evolutionary innovation requires that protein function can change with only a few mutations.”
The paper relies on this without support.
Proteins exist in physical space.
“Proteins are objects that exist in physical space.”
The paper relies on this without support.
Many fundamental evolutionary processes that give rise to the diversity of protein sequences observed today are understood.
“Similarly, we understand many of the fundamental evolutionary processes that give rise to the diversity of protein sequences observed today.”
The paper relies on this without support.
Physics and evolution are key properties of proteins.
“These two elements, physics and evolution, are the key properties of proteins and our models might benefit from being structured explicitly to incorporate evolutionary and physics-based inductive biases.”
The paper relies on this without support.
Biological sequences are not natural language.
“Biological sequences are not natural language”
The paper relies on this without support.
Filtering sequences based on percent identity is a conventional crude address to the phylogenetic assumption problem.
“Conventionally, this problem is crudely addressed by filtering sequences based on percent identity.”
The paper relies on this without support.
At the upper limit of large natural protein databases, larger models continue to yield improved performance.
“At the upper limit of large natural protein databases, larger models continue to yield improved performance.”
The paper relies on this without support.
For individual protein families or application-specific protein datasets, brute force hyperparameter search is the gold standard for selecting model architectures and number of parameters.
“However, for individual protein families or other application specific protein data sets, the gold standard is to select model architectures and number of parameters via brute force hyperparameter search methods.”
The paper relies on this without support.
Transformers scale quadratically with sequence length.
“Transformers scale quadratically with sequence length”
The paper relies on this without support.
Recurrent neural networks scale linearly with sequence length.
“This problem is less extreme for recurrent neural networks, which scale linearly with sequence length”
The paper relies on this without support.
Increasing model size, compute power, and dataset size will continue to improve performance of protein language models.
“Increasing model size, compute power, and dataset size will only continue to improve performance of protein language models.”
The paper relies on this without support.
Protein language models are transforming computational protein biology today due to their ease of use and widespread applicability.
“Already, these methods are transforming computational protein biology today due to their ease of use and widespread applicability.”
The paper relies on this without support.
Augmenting language models with protein-specific properties such as structure and function offers an already successful route towards richer representations and novel biology.
“Furthermore, augmenting language models with protein specific properties such as structure and function offers one already successful route towards even richer representations and novel biology.”
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?
In the citing paper
“Instead of using multi-headed self-attention (MHSA), we adopt the Gated Attention Unit (GAU) (line 8 in Algorithm 1), which has shown great promise as an alternative to MHSAs with smaller memory consumption and faster convergence rate (25).”
From High-resolution de novo structure prediction from primary sequence (2022) · cites this paper as [25]
In the citing paper
“The original GAU in (25) also suggests that relu 2 (•) performs better compared to softmax in terms of both computation speed and convergence rate.”
From High-resolution de novo structure prediction from primary sequence (2022) · cites this paper as [25]
Citations
Cites
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017) · arXiv · doi:10.48550/arxiv.1706.03762 · Cited by 120,404
Hochreiter, S., & Schmidhuber, J. (1997) · Neural Computation · doi:10.1162/neco.1997.9.8.1735 · Cited by 102,318
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2018) · arXiv · doi:10.48550/arxiv.1810.04805 · Cited by 73,173
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019) · Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) · doi:10.18653/v1/n19-1423 · Cited by 54,325
Berman, H. M. (2000) · Nucleic Acids Research · doi:10.1093/nar/28.1.235 · Cited by 40,196
ImageNet Large Scale Visual Recognition Challenge
Russakovsky, O., Deng, J., Su, H., Krause, J., Satheesh, S., Ma, S., Huang, Z., Karpathy, A., Khosla, A., Bernstein, M., et al. (2015) · International Journal of Computer Vision · doi:10.1007/s11263-015-0816-y · Cited by 39,380
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
MRBAYES: Bayesian inference of phylogenetic trees.
Huelsenbeck, J. P., & Ronquist, F. (2001) · Bioinformatics · doi:10.1093/bioinformatics/17.8.754 · Cited by 21,937
GROMACS 4: Algorithms for Highly Efficient, Load-Balanced, and Scalable Molecular Simulation
Hess, B., Kutzner, C., van der Spoel, D., & Lindahl, E. R. (2008) · Journal of Chemical Theory and Computation · doi:10.1021/ct700301q · Cited by 15,830
Flagged first, then by citations
3 of the 133 references couldn’t be matched to a known paper.
Research works printed but not linked: 4
- Radford, 2019, Language models are unsupervised multitask learners, OpenAI blog, 1
- Remmert, 2011, HHblits: lightning-fast iterative protein sequence searching by HMM-HMM alignment, Nat. Methods, 9, 173, 10.1038/nmeth.1818 · DOI · Paper page
- Remmert, 2011, HHblits: lightning-fast iterative protein sequence searching by HMM-HMM alignment, Nat. Methods, 9, 173, 10.1038/nmeth.1818 · DOI · Paper page
- Tsirigos, 2015, The TOPCONS web server for consensus prediction of membrane protein topology and signal peptides, Nucleic Acids Res., 43, W401-7, 10.1093/nar/gkv485 · DOI
Cited by
Sorted by citations
Items 401–486 of 486
Ramya, L., Ragunathan, P. T., & Yennamalli, R. M. (2024) · Elsevier eBooks · doi:10.1016/b978-0-323-95502-7.00103-2
ProtParts, an automated web server for clustering and partitioning protein datasetPreprint
Li, Y., & Barra, C. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.07.12.603234
“The capacity of accurately predicting protein structure and function using machine learning has largely accelerated the identification of sequence patterns and enrichment of the protein universe (3).”
BackgroundHow deep can we decipher protein evolution with deep learning models
Fu, X. (2024) · Patterns · doi:10.1016/j.patter.2024.101043
Boosting Protein Language Models with Negative Sample Mining
Xu, Y., Zhao, X., Song, X., Wang, B., & Yu, T. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-70381-2_13
Deep Video Analysis for Bacteria Genotype PredictionPreprint
Dabouei, A., Mishra, I., Kapur, K., Cao, C., Bridges, A. A., & Xu, M. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.09.16.613253
“Transformers have revolutionized the exploration of longer sequences [44, 45], providing essential insights in evolutionary biology.”
BackgroundUncovering the domain language of protein function and protein networks using DANSyPreprint
Shimpi, A. A., & Naegle, K. M. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.12.04.626803
“Insights from broad surveys of both protein domain architectures and amino acid sequences suggest that proteins operate with sets of rules akin to grammar within natural languages [10, 13, 14].”
HieVi: Protein Large Language Model for proteome-based phage clusteringPreprint
Panigrahi, S., Ansaldi, M., & Ginet, N. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.12.17.627486
A Prompt and Contrastive Learning-Based Model for Plant Antimicrobial Peptide Prediction
Qin, Z., Meng, J., Li, H., Tang, Y., & Luan, Y. (2025) · Communications in computer and information science · doi:10.1007/978-981-96-1907-8_14
Saha, D., Borah, N. J., & Mamud, A. (2025) · Elsevier eBooks · doi:10.1016/b978-0-443-31550-3.00014-7
Interdisciplinary Approaches to Leverage Biomarker Discovery for Cancer Treatment
Khatami, F., & Rezaei, N. (2025) · Interdisciplinary cancer research · doi:10.1007/16833_2025_497
Mitchell, J. (2025) · Queensland University of Technology · doi:10.5204/thesis.eprints.256582
Khatri, K., Levy, R. M., & Haldane, A. (2025) · PubMed
Varghese, D. M., Athulya, T., Mohani, V. K., & Ahmad, S. (2025) · Methods in molecular biology · doi:10.1007/978-1-0716-4623-6_13
Predicting the Pathogenicity of Human Protein Variants: Not Only a Matter of Residue Labeling
Manfredi, M., Vazzana, G., Babbi, G., Bertolini, E., Savojardo, C., Martelli, P. L., & Casadio, R. (2025) · Methods in molecular biology · doi:10.1007/978-1-0716-4623-6_12
Exploring ProtFlash: An Efficient Approach to Protein Data Analysis
Wang, L., Xue, Z., & Wang, Y. (2025) · Methods in molecular biology · doi:10.1007/978-1-0716-4623-6_3
Denger, A., & Helms, V. (2025) · Molecules · doi:10.3390/molecules30153226 · Cited by 1
Papadopoulos, F., Sánchez-Elsner, T., Niranjan, M., & Heinson, A. (2025) · PLoS ONE · doi:10.1371/journal.pone.0325531
“Natural Language Processing (NLP) has seen several new Self-Supervised Learning (SSL) algorithms [5].”
InfluentialHuman Symbolic Evolution: A 7E Cognition Approach
Gontier, N. (2025) · Elsevier eBooks · doi:10.1016/b978-0-323-95504-1.00863-2
Uncovering differential tolerance to deletions versus substitutions with a protein language model
Goldman, G., Chati, P., & Ntranos, V. (2025) · Cell Systems · doi:10.1016/j.cels.2025.101373
Generative models for drug design
Zhou, Y., & Wang, Y. (2025) · Elsevier eBooks · doi:10.1016/b978-0-44-332908-1.00015-5
Protein Language Model–Aligned Spectra Embeddings for De Novo Peptide SequencingPreprint
Naderializadeh, N., Dallago, C., Soderblom, E. J., & Soderling, S. H. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.10.01.679857
“In this work, we leverage protein language models (PLMs) [3, 46, 11, 24, 17] to enhance the training procedure of deep learning de novo peptide sequencing pipelines.”
Yang, K., Chen, Y., Wei, Y., Xiang, M., Zhuo, L., Fu, X., Cao, D., & Zhang, W. (2025) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.5c01907
Meng, D., Alvarez, H. M. G., Glavina, J., Leonetti, C. O., Pollastri, G., & Chemes, L. B. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.10.29.685455 · Cited by 1
Ma, R., He, C., Zhang, Z., Zheng, H., & Duan, L. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.11.11.687944
Harrigan, W., Ferrell, B. D., Wommack, K. E., Polson, S. W., Schreiber, Z. D., & Belcaid, M. (2024) · Figshare · doi:10.6084/m9.figshare.25698243
Additional file 1 of Protein sequence analysis in the context of drug repurposing
Sánchez, N. G., Carro, E. U., Prieto-Santamaría, L., & Rodríguez‐González, A. (2024) · Figshare · doi:10.6084/m9.figshare.26719884
Additional file 1 of A-Prot: protein structure modeling using MSA transformer
Hong, Y., Lee, J., & Ko, J. (2022) · Figshare · doi:10.6084/m9.figshare.19373539
Additional file 1 of TMbed: transmembrane proteins predicted through language model embeddings
Bernhofer, M., & Rost, B. (2022) · Figshare · doi:10.6084/m9.figshare.20454488
Tan, Y., Zhou, X., Yu, Z., Xiao, Q., Li, X., & Xu, Y. (2025) · Data Mining and Knowledge Discovery · doi:10.1007/s10618-025-01179-3
“In the biomedical field, representation learning has the potential to enhance our comprehension of gene expression data and protein interaction networks, thereby facilitating research on disease mechanisms and the development of novel therapeutic approaches (Bepler and Berger 2021).”
SenSeqNet : A Deep Learning Framework for Cellular Senescence Detection From Protein Sequences
Jiang, H., Deng, D., Yuan, Y. L., Ren, J., Yang, X., Liu, S., Tan, B., Lin, L., & Lubin, L. I. U. (2025) · Aging Cell · doi:10.1111/acel.70344 · Cited by 1
“sequences are of particular interest, as they encode the structural and functional properties of proteins that mediate nearly all cellular processes (Bepler and Berger 2021).”
Machine Learning for Sequence‐to‐Function Approaches
Barghout, R. A., Kirby, M., Zheng, A., Chinas, L., Mohammadi, M., Xu, Z., Sánchez-Lengeling, B., & Mahadevan, R. (2026) · Wiley eBooks · doi:10.1002/9783527850532.ch4 · Cited by 1
Better Inputs, Better Learning: A Peptide Embedding Tutorial for Proteomic Mass Spectrometry
Squires, L., Chavez, J. H. G., Nilsson, A., Käll, L., & Payne, S. H. (2026) · Journal of Proteome Research · doi:10.1021/acs.jproteome.5c00563
Context-Aware Protein Representations Using Protein Language Models and Optimal TransportPreprint
Patel, S. S., & Naderializadeh, N. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.01.24.701517
“The most common method to aggregate these token-level embeddings into a single representation for the entire protein is average pooling [2, 52, 53, 58].”
Papkiauri, A., Urushadze, L., Tevdoradze, T., Sidamonidze, K., Tomashvili, G., Gavashelidze, M., Ninua, L., Daraselia, I., Kiknavelidze, S., Melikishvili, N., et al. (2026) · Frontiers in Microbiology · doi:10.3389/fmicb.2026.1735728
Bouvier, B. (2026) · Journal of Chemical Theory and Computation · doi:10.1021/acs.jctc.5c01690
EvoPool: Evolution-Guided Pooling of Protein Language Model EmbeddingsPreprint
Naderializadeh, N., & Singh, R. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.02.02.703349 · Cited by 2
“To obtain fixed-size protein-level representations from residue-level embeddings produced by PLMs, most prior work has relied on simple aggregation operations, most notably average pooling [29–37].”
Leveraging learned representations and multitask learning for lysine methylation site discovery
Charih, F., Boulter, M., Biggar, K. K., & Green, J. R. (2026) · Scientific Reports · doi:10.1038/s41598-026-39136-9
“…have shown that multitask language models better capture the semantic organization of proteins by training a bidirectional long short-term memory (LSTM) to complete three tasks simultaneously: masked language modeling, residue-residue contact prediction, and structural similarity prediction [16].”
LiBRe: A Ligand-Aware Sequence-Based Binding Residue Prediction Model for Virtual Screening
Kang, K., Kim, M., Kim, J., Sel, S., & Song, G. (2026) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.5c02883
“Building on previous studies that employed protein language models for residue representation, 46,47 which showed clear advantages over traditional handcrafted features, we also adopted this strategy and encoded each residue using the tokenizer of ESM-2.”
TraSPIN: inferring transporter specificity through language model-guided graph learning
Pan, T., Liu, J., Xu, J., Han, R. D., Ran, Z., Guo, X., Hao, Y., Bi, Y., Li, F., & Song, J. (2026) · Genome biology · doi:10.1186/s13059-026-04309-2
A DNN Biophysics Model with Topological and Electrostatic Features
Sliheet, E., Talha, M. A., & Geng, W. (2026) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.5c03104
“…of Machine Learning (ML)-based protein prediction focuses on using amino acid sequences to predict protein structure or structure-related properties, resulting in the trending protein language models (PLMs), 7,8 which is a transformer-based model inspired by Natural Language Processing (NLP).”
Recent advances in multimodal-driven protein AI
Yang, Y., Huang, H., Zhang, J., Li, W., & Wang, W. (2026) · Biophysics Reviews · doi:10.1063/5.0273453
Behbahani, Y. M. (2022) · theses.fr (ABES) · doi:10.70675/bfdded7czb638z455azaedaz4ae9f8d0a13d
From sequences to knowledge, improving and learning from sequence alignments
Blassel, L. (2022) · theses.fr (ABES) · doi:10.70675/edb2db56z9b8ez4f44zb76az28681bf8bc51
Shao, J., Chen, S., Wang, Z., Chen, Z., & Liu, B. (2026) · Genome Research · doi:10.1101/gr.280816.125
Protein Design Enters the Artificial Intelligence Era: Foundations, Tools, and Emerging Paradigms
Mi, Y., Shukla, A., Tangney, M., Tabirca, S., & Yallapragada, V. V. B. (2026) · Computational and Structural Biotechnology Journal · doi:10.34133/csbj.0105
“Additionally, contrastive learning between sequence and structure representations has proven particularly effective in developing rich, informative protein embeddings [ 8 , 32 ].”
Maher, V., Martin, D., Spetzler, D. B., Zhao, Z., O’Neill, H. A., & Woodbury, N. W. (2026) · Bioinformatics Advances · doi:10.1093/bioadv/vbag118
“This approach is similar to one that has been used previously to successfully generate protein sequence embeddings (Bepler and Berger 2021).”
Structure-aware protein function prediction at isoform resolutionPreprint
Jiang, F., Zhao, R., Liang, F., Zhao, X., Cui, T., Tang, Z., Zhang, Y. S., Chen, D., Xu, M., Shuai, Y., et al. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.04.24.720502
Structure-derived synthetic sequences guide a protein language model toward metalloproteinsPreprint
Peteani, G., Sgueglia, G., Lemmin, T., & Chino, M. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.04.30.722007
Wang, X., Feng, X., Tarafder, S., & Bhattacharya, D. (2026) · Nucleic Acids Research · doi:10.1093/nar/gkag425
Simple baselines rival protein language models in dense combinatorial protein landscapesPreprint
Talpir, I., & Fleishman, S. J. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.05.01.722313 · Cited by 2
CodonMoE: DNA language models for codon-dependent mRNA prediction
Du, S., Liang, L., Li, J., & Kingsford, C. (2026) · Bioinformatics · doi:10.1093/bioinformatics/btag285
“These models aim to capture the complex patterns and dependencies within genomic sequences, much like how general LLMs learn the intricacies of human language (Bepler and Berger 2021).”
Jiang, T., Zhu, Y., Liu, Z., Yi, X., Zhang, Q., & Song, S. (2026) · Molecular Genetics and Genomics · doi:10.1007/s00438-026-02433-z
“The results demonstrate that sequence, structural, and Gene Ontology functional semantic modalities encode complementary biological information (Bepler and Berger 2021; Nguyen and Hy 2024).”
Beyond profiles: supervised repeat annotation using protein embeddingsPreprint
Qiu, K., Ludwiczak, J., Lupas, A. N., & Dunin-Horkawicz, S. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.05.19.725729
Liebeskind, K., Francklyn, C. S., & Barrantes-Reynolds, R. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.05.22.727252
Learning Protein-Protein Interaction Interfaces from Multimodal RepresentationsPreprint
Gardinazzi, Y., Garcia, E. N. V., Senci, S., Vora, D. D., Feltrin, A., & Cuturello, F. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.05.29.728739
From prediction to discovery: AI as an observatory of physical organization in protein space
Zheng, Y., Zhang, Z., Wang, Y., Kang, W., Ren, W., & Tang, Q.-Y. (2026) · AI for Science · doi:10.1088/3050-287x/ae78ea
“PLMs extend this logic from pairwise couplings to higher-order context dependence, and their likelihoods, embeddings, and mutation scores therefore reflect not fundamental physical laws in isolation, but the statistical imprint of physical viability sieved through evolution [50, 126].”
Zhang, S., Qin, Y., Zhu, H., Meng, F., Jia, L., & Zheng, X. (2026) · Briefings in Bioinformatics · doi:10.1093/bib/bbag279
Wei, M., Bai, S., Chen, W., Wu, J., Pan, Y., Zhao, W., Qian, W., Cao, J., Pan, G., Bai, X., et al. (2026) · Phenomics · doi:10.1007/s43657-025-00295-w
“Language models such as recurrent neural networks (RNN), Transformers, and their variants have recently emerged as powerful approaches for sequence generation and for deriving “content-aware” representations from extensive sequence datasets (Bepler and Berger 2021).”
DeepCDS: Ab initio coding sequence prediction in prokaryotic short nucleotide sequencesPreprint
Nielsen, L. S., Nielsen, H., & Winther, O. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.06.17.732633
“…sequence-based features, protein language models (pLMs) such as ESM-2 [26], ESM C [27], and ProtT5 [28] have substantially enhanced protein sequence representations by learning evolutionary, structural, and functional dependencies across large corpora of unlabeled protein sequences [26, 28, 29].”
Integrating experimental and computational approaches for protein–protein interaction discovery
Rapti, A., Janssen, B. J. C., & Bonvin, A. M. J. J. (2026) · Nature Reviews Bioengineering · doi:10.1038/s44222-026-00464-0
“Although pLMs do not use structural information as input, their embeddings implicitly capture structural and evolutionary constraints, enabling alignment-free protein modelling without the need for multiple sequence alignments 62,63 .”
Rannon, E., & Burstein, D. (2026) · Bioinformatics · doi:10.1093/bioinformatics/btag226 · Cited by 1
“The ability of language models (LMs) to learn complex patterns from large unlabeled corpora makes them well-suited to address this challenge (Bepler and Berger 2021, Iuchi et al. 2021, Ofer et al. 2021, Zhang et al. 2023, Oikonomou et al. 2024, Rannon and Burstein 2025).”
Luo, W., Liu, Z., Wu, H., Xiao, X., Lin, W., & Qiu, W. (2026) · Current Genomics · doi:10.1016/j.curgen.2026.07.002
Hany, A., Dahroug, A., & Moussa, S. M. (2026) · IEEE International Conference on Smart Sustainable Systems for Computer and Engineering Applications (3SCEA) · doi:10.1109/3scea68071.2026.11603102
“In [25], a "protein language" for evolution was presented, but they failed to integrate these functional priors with the raw nucleotide-level "grammar" found in long-range DNA.”
Chemical Language Model Applications in Medicinal Chemistry
Mishra, G., & Dixit, V. A. (2026) · Royal Society of Chemistry eBooks · doi:10.1039/9781837070206-00543
Endoh, T., Camer, G. A., Kayama, K., Endoh, D., & Teraoka, H. (2026) · Proteomes · doi:10.3390/proteomes14030036
“Recent studies on protein language models similarly demonstrate that embedding representations encode structural and functional information not reducible to pairwise sequence identity or alignment score [15–17].”
Beyond the parasite: reframing cutaneous leishmaniasis as a host-microbiome-vector ecosystem
Huët, M. A. L., & Bhaw‐Luximon, A. (2026) · International Journal of Pharmaceutics · doi:10.1016/j.ijpharm.2026.127250
Zhang, B., Li, R., Yin, K., Yang, Y., Zhang, J., Jiang, M., Wang, H., Li, S., & Jia, L. (2026) · Briefings in Bioinformatics · doi:10.1093/bib/bbag424
“However, integrating PLM embeddings with established physicochemical and compositional descriptors remains challenging [ 25– 28]: naive concatenation of modalities with vastly different dimensionalities risks allowing higher-dimensional features to dominate biologically informative contributions.”
Singh, D. (2026) · Journal of Proteins and Proteomics · doi:10.1007/s42485-026-00212-3
“…of generative structural design: 20 or fewer macrocycles were tested per target across four proteins, binders were obtained for all four, selected affinities were below 10 nM, and three crystal complexes agreed with design models at Cα RMSD values of 0.7, 1.2, and 1.4 Å (Bepler and Berger 2021).”
InfluentialLiverDCP: A Disease–Cell–Protein Framework for Multi-scale Modeling of Disease BiologyPreprint
Shi, Z., Song, Z., Steveson-Lerner, H., Dong, B., & Zhao, H. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.08.07.743628 · Cited by 1
“Nevertheless, existing approaches remain limited in their ability to jointly model protein interaction topology, disease-state-specific network rewiring, and intrinsic protein properties within a unified framework 35 .”
Tokenization in Protein Language Models: Methods, Taxonomy, and Applications
Linoubli, A., Abidi, K., & Farah, M. (2026) · Communications in computer and information science · doi:10.1007/978-981-92-2597-2_11
Shoaib, M., & Waqas, A. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.08.26.747234
Deep Learning in Enzyme Function Prediction and Novel Enzyme Discovery
Gao, R., Liu, Y., Duan, C., Qin, N., Liu, K., Tsuboi, T., & Chen, Z. (2026) · ACS Synthetic Biology · doi:10.1021/acssynbio.6c00633
Long, X., Lin, Y., Xie, J., Wang, J., Wei, X., Wei, L., Deng, Z., & Gao, J. (2026) · Wiley Interdisciplinary Reviews Computational Molecular Science · doi:10.1002/wcms.70087
“The evolution-structure– function link in protein language explains why curated datasets are so valuable; they capture the intrinsic biological patterns that algorithms, whether simple or complex, need to learn, which justifies the priority of data curation over architectural complexity [115].”
AmyloDeep: pLM-based ensemble model for predicting amyloid propensity from the amino acid sequence
Davtyan, A., Khachatryan, A., & Petrosyan, R. A. (2026) · Bioinformatics Advances · doi:10.1093/bioadv/vbag296
Mishra, M. P., Tripathy, P. K., & Tripathy, C. (2026) · International Conference on Advancements in Smart, Secure and Intelligent Computing (ASSIC) · doi:10.1109/assic70037.2026.11716787
“The method makes use of Bepler and Berger’s pre-trained language model [25] to obtain protein embeddings.”
Wang, Z., Zhang, Z., Yang, Y., Qu, Y., Mu, Y., Wang, S., 韩楠玉, Zheng, L., & Li, W. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.10.05.756918
“In addition to structural information, protein evolution also encodes rich functional constraints [36, 37, 38, 39].”
Zhang, J., Zhang, J., Zheng, L., Ruan, H., Li, Y., Li, Y., & Luo, G. (2026) · Journal of Molecular Graphics and Modelling · doi:10.1016/j.jmgm.2026.109610
Cheng, H., Sun, C., Ye, C., Wu, C., & Zhao, D. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.10.07.757219
Introduction to the principles of protein engineering
Sharp, J. F., Nelson, M. K., Rustembek, M., Samanta, R., & Stern, L. A. (2026) · Elsevier eBooks · doi:10.1016/b978-0-443-44536-1.00010-9
Vijayanand, M., & Patil, K. K. (2026) · International Journal of Peptide Research and Therapeutics · doi:10.1007/s10989-026-10870-y
“Protein language models (pLMs) treat amino acid sequences as a language and learn contextual representations from large sequence corpora, capturing evolutionary and physicochemical information without explicit structural input (Bepler and Berger 2021; Heinzinger et al. 2024).”
Towards Automated Protein Family Annotation in SCOPe Using Protein Language Models
Traseira, M., & Abbass, J. (2026) · Bioinformatics and Biology Insights · doi:10.1177/11779322261501059
FAIRsharing Team, & Morgat, A. (2015) · FAIRsharing.org · doi:10.25504/fairsharing.pn1sr5
Bhuiyan, M. N. I., & Hossain, S. J. (2026) · Environmental Microbiology Reports · doi:10.1111/1758-2229.70454
“By learning statistical patterns in protein sequences, these methods can support functional annotation, identify candidate proteins, and prioritise sequences for experimental investigation (Bepler and Berger 2021; Bileschi et al. 2022; Chandra et al. 2023).”
Metadata
- Authors
- Bepler, T., & Berger, B.
- Affiliations
- New York Structural Biology Center (US)
- MIT Computer Science and Artificial Intelligence Laboratory (US)
- Corresponding author
- Bepler, Tristan
- Berger, Bonnie
- Venue
- Cell Systems
- Publisher
- Elsevier BV
- Published
- 2021-06
- Volume
- 12
- Issue
- 6
- Pages
- 654-669.e3
- ISSN
- 2405-4712 (ISSN-L)
- PMC
- PMC8238390
- OpenAlex
- W3166142427
- Cited by
- 483
- References
- 133
- Citations by year
- 2026: 87 · 2025: 120 · 2024: 127 · 2023: 82 · 2022: 55 · 2021: 12
- FWCI
- 21.74
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- Locations
- Cell Systems (publishedVersion, cc-by-nc-nd)
- PubMed (publishedVersion)
- PubMed Central (acceptedVersion)
- DSpace@MIT (Massachusetts Institute of Technology) (submittedVersion, cc-by-nc-nd)
- Funders
- NIH (R35 GM141861)
- Simons Foundation (SF349247)
- Topics
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Machine Learning in Bioinformatics
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Protein Structure and Dynamics
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Fractal and DNA sequence analysis
- Keywords
- protein language models, protein sequence representation, protein structure prediction, protein function prediction, knowledge distillation, transfer learning, protein design, therapeutic design
- MeSH terms
- Machine Learning, Amino Acid Sequence, Language, Proteins, Databases, Protein
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 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
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
