Comparative evaluation of AlphaFold2 and disorder predictors for prediction of intrinsic disorder, disorder content and fully disordered proteins

Zhao, Bi; Ghadermarzi, Sina; Kurgan, Lukasz

doi:10.1016/j.csbj.2023.06.001

Cited by 11 publications

(7 citation statements)

References 113 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In line with our analysis, a recent study by Peng and Zhao 76 further corroborates that many high-pLDDT predictions for Drosophila de novo proteins exhibit instability during molecular dynamics simulations under realistic conditions. This finding lends additional support to the notion that several regions characterized by high pLDDT are also predicted to be disordered.…”

Section: Discussionsupporting

confidence: 90%

See 1 more Smart Citation

Random, de novo, and conserved proteins: How structure and disorder predictors perform differently

Middendorf,

Eicholt

2024

Proteins

View full text Add to dashboard Cite

Understanding the emergence and structural characteristics of de novo and random proteins is crucial for unraveling protein evolution and designing novel enzymes. However, experimental determination of their structures remains challenging. Recent advancements in protein structure prediction, particularly with AlphaFold2 (AF2), have expanded our knowledge of protein structures, but their applicability to de novo and random proteins is unclear. In this study, we investigate the structural predictions and confidence scores of AF2 and protein language model‐based predictor ESMFold for de novo and conserved proteins from Drosophila and a dataset of comparable random proteins. We find that the structural predictions for de novo and random proteins differ significantly from conserved proteins. Interestingly, a positive correlation between disorder and confidence scores (pLDDT) is observed for de novo and random proteins, in contrast to the negative correlation observed for conserved proteins. Furthermore, the performance of structure predictors for de novo and random proteins is hampered by the lack of sequence identity. We also observe fluctuating median predicted disorder among different sequence length quartiles for random proteins, suggesting an influence of sequence length on disorder predictions. In conclusion, while structure predictors provide initial insights into the structural composition of de novo and random proteins, their accuracy and applicability to such proteins remain limited. Experimental determination of their structures is necessary for a comprehensive understanding. The positive correlation between disorder and pLDDT could imply a potential for conditional folding and transient binding interactions of de novo and random proteins.

show abstract

Section: Discussionsupporting

confidence: 90%

“…Consequently, when dealing with random and de novo proteins, employment of disorder and structure predictors combined with molecular dynamics simulation, 76 becomes imperative for comprehensive assessment of their structural properties. It is essential to note that our research exclusively focused on de novo proteins derived from Drosophila.…”

Section: Discussionmentioning

confidence: 99%

Random, de novo, and conserved proteins: How structure and disorder predictors perform differently

Middendorf,

Eicholt

2024

Proteins

View full text Add to dashboard Cite

show abstract

“…In the case of analyzed L. pneumophila effector models, we estimated IDRs to be present in at least 24% of these predicted effector models, as compared to 4.8% estimated for the remainder of this bacterium’s proteome. While these estimates are based on sequence-based IDR predictions, they are further supported or even extended by AlphaFold2 predictions, which are accepted to have high accuracy for the identification of such regions (Zhao et al , 2023). Notably, 8 L. pneumophila effector models do not contain any recognizable structural elements, suggesting these proteins to be full-length IDRs.…”

Section: Discussionmentioning

confidence: 99%

Global atlas of predicted functional domains inLegionella pneumophilaDot/Icm translocated effectors

Patel,

Stogios,

Jaroszewski

et al. 2024

Preprint

View full text Add to dashboard Cite

Legionella pneumophila utilizes the Dot/Icm type IVB secretion system to deliver hundreds of effector proteins inside eukaryotic cells to ensure intracellular replication. Our understanding of the molecular functions of the largest pathogenic arsenal known to the bacterial world remains incomplete. By leveraging advancements in 3D protein structure prediction, we provide a comprehensive structural analysis of 368 L. pneumophila effectors, representing a global atlas of predicted functional domains summarized in a database (https://pathogens3d.org/legionella-pneumophila). Our analysis identified 157 types of diverse functional domains in 287 effectors, including 159 effectors with no prior functional annotations. Furthermore, we identified 35 unique domains in 30 effector models that have no similarity with experimentally structurally characterized proteins, thus, hinting at novel functionalities. Using this analysis, we demonstrate the activity of thirteen domains, including three unique folds, predicted in L. pneumophila effectors to cause growth defects in the Saccharomyces cerevisiae model system. This illustrates an emerging strategy of exploring synergies between predictions and targeted experimental approaches in elucidating novel effector activities involved in infection.

show abstract

“…We evaluate these predictions by comparing these propensities against the ground truth annotations of DLs. We rely on the same metrics that were used in the CAID1, CAID2, and other recent assessments of disorder predictors [ 28 , 31 , 33 , 56 , 57 , 58 , 59 , 60 ]. They include AUC, lowAUCratio, and Area Under the Precision-Recall Curve (AUPRC), which quantify the performance of the predicted propensities.…”

Section: Methodsmentioning

confidence: 99%

“…We use the Anderson–Darling test at 0.05 significance to test whether the resulting measurements are normal; for normal measurements, we assess significance using the t -test; otherwise, we utilize the Wilcoxon test. While such tests are commonly used in disorder prediction assessments [ 58 , 59 , 60 ], they were not reported in CAID2, where only the overall/dataset-level scores are available.…”

Section: Methodsmentioning

confidence: 99%

Assessment of Disordered Linker Predictions in the CAID2 Experiment

Wang,

Hu,

et al. 2024

Biomolecules

Self Cite

View full text Add to dashboard Cite

Disordered linkers (DLs) are intrinsically disordered regions that facilitate movement between adjacent functional regions/domains, contributing to many key cellular functions. The recently completed second Critical Assessments of protein Intrinsic Disorder prediction (CAID2) experiment evaluated DL predictions by considering a rather narrow scenario when predicting 40 proteins that are already known to have DLs. We expand this evaluation by using a much larger set of nearly 350 test proteins from CAID2 and by investigating three distinct scenarios: (1) prediction residues in DLs vs. in non-DL regions (typical use of DL predictors); (2) prediction of residues in DLs vs. other disordered residues (to evaluate whether predictors can differentiate residues in DLs from other types of intrinsically disordered residues); and (3) prediction of proteins harboring DLs. We find that several methods provide relatively accurate predictions of DLs in the first scenario. However, only one method, APOD, accurately identifies DLs among other types of disordered residues (scenario 2) and predicts proteins harboring DLs (scenario 3). We also find that APOD’s predictive performance is modest, motivating further research into the development of new and more accurate DL predictors. We note that these efforts will benefit from a growing amount of training data and the availability of sophisticated deep network models and emphasize that future methods should provide accurate results across the three scenarios.

show abstract

Comparative evaluation of AlphaFold2 and disorder predictors for prediction of intrinsic disorder, disorder content and fully disordered proteins

Cited by 11 publications

References 113 publications

Random, de novo, and conserved proteins: How structure and disorder predictors perform differently

Random, de novo, and conserved proteins: How structure and disorder predictors perform differently

Global atlas of predicted functional domains inLegionella pneumophilaDot/Icm translocated effectors

Assessment of Disordered Linker Predictions in the CAID2 Experiment

Contact Info

Product

Resources

About