SignalP: The Evolution of a Web Server.

SignalP ( https://services.healthtech.dtu.dk/services/SignalP-6.0/ ) is a very popular prediction method for signal peptides, the intrinsic signals that make proteins secretory. The SignalP web server has existed since 1995 and is now in its sixth major version. In this historical account, we (three authors who have taken part in the entire journey plus the first author of the latest version) describe the differences between the versions and discuss the various decisions taken along the way.

Dipeptidyl peptidases and E3 ligases of N-degron pathways cooperate to regulate protein stability.

N-degrons are short sequences located at protein N-terminus that mediate the interaction of E3 ligases (E3s) with substrates to promote their proteolysis. It is well established that N-degrons can be exposed following protease cleavage to allow recognition by E3s. However, our knowledge regarding how proteases and E3s cooperate in protein quality control mechanisms remains minimal. Using a systematic approach to monitor the protein stability of an N-terminome library, we found that proline residue at the third N-terminal position (hereafter "P+3") promotes instability. Genetic perturbations identified the dipeptidyl peptidases DPP8 and DPP9 and the primary E3s of N-degron pathways, UBR proteins, as regulators of P+3 bearing substrate turnover. Interestingly, P+3 UBR substrates are significantly enriched for secretory proteins. We found that secretory proteins relying on a signal peptide (SP) for their targeting contain a "built-in" N-degron within their SP. This degron becomes exposed by DPP8/9 upon translocation failure to the designated compartments, thus enabling clearance of mislocalized proteins by UBRs to maintain proteostasis.

Structural and functional insights into the C-terminal signal domain of the Bacteroidetes type-IX secretion system.

Gram-negative bacteria from the Bacteroidota phylum possess a type-IX secretion system (T9SS) for protein secretion, which requires cargoes to have a C-terminal domain (CTD). Structurally analysed CTDs are from Porphyromonas gingivalis proteins RgpB, HBP35, PorU and PorZ, which share a compact immunoglobulin-like antiparallel 3+4 ß-sandwich (ß1-ß7). This architecture is essential as a P. gingivalis strain with a single-point mutant of RgpB disrupting the interaction of the CTD with its preceding domain prevented secretion of the protein. Next, we identified the C-terminus ('motif C-t.') and the loop connecting strands ß3 and ß4 ('motif Lß3ß4') as conserved. We generated two strains with insertion and replacement mutants of PorU, as well as three strains with ablation and point mutants of RgpB, which revealed both motifs to be relevant for T9SS function. Furthermore, we determined the crystal structure of the CTD of mirolase, a cargo of the Tannerella forsythia T9SS, which shares the same general topology as in Porphyromonas CTDs. However, motif Lß3ß4 was not conserved. Consistently, P. gingivalis could not properly secrete a chimaeric protein with the CTD of peptidylarginine deiminase replaced with this foreign CTD. Thus, the incompatibility of the CTDs between these species prevents potential interference between their T9SSs.

Identification of Auxiliary Organellar Targeting Signals for Plant Peroxisomes Using Bioinformatic Analysis of Large Protein Sequence Datasets Followed by Experimental Validation.

Eukaryotic cells are compartmentalized by membrane-bounded organelles to ensure that specific biochemical reactions and cellular functions occur in a spatially restricted manner. The subcellular localization of proteins is largely determined by their intrinsic targeting signals, which are mainly constituted by short peptides. A complete organelle targeting signal may contain a core signal (CoreS) as well as auxiliary signals (AuxiS). However, the AuxiS is often not as well characterized as the CoreS. Peroxisomes house many key steps in photorespiration, besides other crucial functions in plants. Peroxisome targeting signal type 1 (PTS1), which is carried by most peroxisome matrix proteins, was initially recognized as a C-terminal tripeptide with a "canonical" consensus of [S/A]-[K/R]-[L/M]. Many studies have shown the existence of auxiliary targeting signals upstream of PTS1, but systematic characterizations are lacking. Here, we designed an analytical strategy to characterize the auxiliary targeting signals for plant peroxisomes using large datasets and statistics followed by experimental validations. This method may also be applied to deciphering the auxiliary targeting signals for other organelles, whose organellar targeting depends on a core peptide with assistance from a nearby auxiliary signal.

Secretion of the cytoplasmic and high molecular weight ß-galactosidase of Paenibacillus wynnii with Bacillus subtilis.

BACKGROUND: The gram-positive bacterium Bacillus subtilis is widely used for industrial enzyme production. Its ability to secrete a wide range of enzymes into the extracellular medium especially facilitates downstream processing since cell disruption is avoided. Although various heterologous enzymes have been successfully secreted with B. subtilis, the secretion of cytoplasmic enzymes with high molecular weight is challenging. Only a few studies report on the secretion of cytoplasmic enzymes with a molecular weight > 100 kDa. RESULTS: In this study, the cytoplasmic and 120 kDa ß-galactosidase of Paenibacillus wynnii (ß-gal-Pw) was expressed and secreted with B. subtilis SCK6. Different strategies were focused on to identify the best secretion conditions. Tailormade codon-optimization of the ß-gal-Pw gene led to an increase in extracellular ß-gal-Pw production. Consequently, the optimized gene was used to test four signal peptides and two promoters in different combinations. Differences in extracellular ß-gal-Pw activity between the recombinant B. subtilis strains were observed with the successful secretion being highly dependent on the specific combination of promoter and signal peptide used. Interestingly, signal peptides of both the general secretory- and the twin-arginine translocation pathway mediated secretion. The highest extracellular activity of 55.2 ± 6 µkat/Lculture was reached when secretion was mediated by the PhoD signal peptide and expression was controlled by the PAprE promoter. Production of extracellular ß-gal-Pw was further enhanced 1.4-fold in a bioreactor cultivation to 77.5 ± 10 µkat/Lculture with secretion efficiencies of more than 80%. CONCLUSION: For the first time, the ß-gal-Pw was efficiently secreted with B. subtilis SCK6, demonstrating the potential of this strain for secretory production of cytoplasmic, high molecular weight enzymes.

Positive charges promote the recognition of proteins by the chaperone SlyD from Escherichia coli.

SlyD is a widely-occurring prokaryotic FKBP-family prolyl isomerase with an additional chaperone domain. Often, such as in Escherichia coli, a third domain is found at its C-terminus that binds nickel and provides it for nickel-enzyme biogenesis. SlyD has been found to bind signal peptides of proteins that are translocated by the Tat pathway, a system for the transport of folded proteins across membranes. Using peptide arrays to analyze these signal peptide interactions, we found that SlyD interacted only with positively charged peptides, with a preference for arginines over lysines, and large hydrophobic residues enhanced binding. Especially a twin-arginine motif was recognized, a pair of highly conserved arginines adjacent to a stretch of hydrophobic residues. Using isothermal titration calorimetry (ITC) with purified SlyD and a signal peptide-containing model Tat substrate, we could show that the wild type twin-arginine signal peptide was bound with higher affinity than an RR>KK mutated variant, confirming that positive charges are recognized by SlyD, with a preference of arginines over lysines. The specific role of negative charges of the chaperone domain surface and of hydrophobic residues in the chaperone active site was further analyzed by ITC of mutated SlyD variants. Our data show that the supposed key hydrophobic residues of the active site are indeed crucial for binding, and that binding is influenced by negative charges on the chaperone domain. Recognition of positive charges is likely achieved by a large negatively charged surface region of the chaperone domain, which is highly conserved although individual positions are variable.

Unlocking the potential of Extensin Signal peptide and Elastin-like polypeptide tag fused to Shigella dysenteriae's IpaDSTxB to improve protein expression and purification in Nicotiana tabacum and Medicagosativa.

Plants are often seen as a potent tool in the recombinant protein production industry. However, unlike bacterial expression, it is not a popular method due to the low yield and difficulty of protein extraction and purification. Therefore, developing a new high efficient and easy to purify platform is crucial. One of the best approaches to make extraction easier is to utilize the Extensin Signal peptide (EXT) to translocate the recombinant protein to the outside of the cell, along with incorporating an Elastin-like polypeptide tag (ELP) to enhance purification and accumulation rates. In this research, we transiently expressed Shigella dysenteriae's IpaDSTxB fused to both NtEXT and ELP in both Nicotiana tabacum and Medicago sativa. Our results demonstrated that N. tabacum, with an average yield of 6.39 ng/µg TSP, outperforms M. sativa, which had an average yield of 3.58 ng/µg TSP. On the other hand, analyzing NtEXT signal peptide indicated that merging EXT to the constructs facilitates translocation of IpaDSTxB to the apoplast by 78.4% and 65.9% in N. tabacum and M. sativa, respectively. Conversely, the mean level for constructs without EXT was below 25% for both plants. Furthermore, investigation into the orientation of ELP showed that merging it to the C-terminal of IpaDSTxB leads to a higher accumulation rate in both N. tabacum and M. sativa by 1.39 and 1.28 times, respectively. It also facilitates purification rate by over 70% in comparison to 20% of the 6His tag. The results show a highly efficient and easy to purify platform for the expression of heterologous proteins in plant.

Residence of the Nucleotide Sugar Transporter Family Members SLC35F1 and SLC35F6 in the Endosomal/Lysosomal Pathway.

The SLC35 (Solute Carrier 35) family members acting as nucleotide sugar transporters are typically localized in the endoplasmic reticulum or Golgi apparatus. It is, therefore, intriguing that some reports document the presence of orphan transporters SLC35F1 and SLC35F6 within the endosomal and lysosomal system. Here, we compared the subcellular distribution of these proteins and found that they are concentrated in separate compartments; i.e., recycling endosomes for SLC35F1 and lysosomes for SLC35F6. Swapping the C-terminal tail of these proteins resulted in a switch of localization, with SLC35F1 being trafficked to lysosomes while SLC35F6 remained in endosomes. This suggested the presence of specific sorting signals in these C-terminal regions. Using site-directed mutagenesis, fluorescence microscopy, and cell surface biotinylation assays, we found that the EQERLL360 signal located in the cytoplasmic tail of human SLC35F6 is involved in its lysosomal sorting (as previously shown for this conserved sequence in mouse SLC35F6), and that SLC35F1 localization in the recycling pathway depends on two YXXΦ-type signals: a Y367KQF sequence facilitates its internalization from the plasma membrane, while a Y392TSL motif prevents its transport to lysosomes, likely by promoting SLC35F1 recycling to the cell surface. Taken together, these results support that some SLC35 members may function at different levels of the endosomal and lysosomal system.

Trial and error: New insights into recombinant expression of membrane-bound insect cytochromes P450 in Escherichia coli systems.

Insect cytochromes P450 (CYP450s) are key enzymes responsible for a wide array of oxidative transformations of both endogenous and exogenous substrates. However, there is currently no a universal guideline established for heterologous expression of membrane-bound CYP450s, which hampers their downstream biochemical and structural studies. In this study, we conducted large-scale screening of protein overexpression in Escherichia coli using 71 insect CYP450 sequences and optimized the expression of a difficult-to-express CYP450 (CYP6HX3) using eight different optimizations, including selection of host strains and expression vectors, alternative of leader signal peptides, and N-terminal modifications. We confirmed that 1) Only insect CYP450s belonging to the CYP347 family could be expressed with N-terminal fusion of ompA2+ signal peptide in E. coli expression system. 2) E. coli Lemo 21 (DE3) effectively improved the expression of CYP6HX3 in the plasma membrane. 3) A brick-red appearance occurred frequently in the expressed thallus or membrane proteins, but this phenomenon could not necessarily indicate successful overexpression of target CYP450s. These findings provide new insights into the recombinant expression of insect CYP450s in E. coli systems and will facilitate the theoretical approaches for functional expression and production of eukaryotic CYP450s.

Signal peptide replacement of Ag43 enables an efficient bacterial cell surface display of receptor-binding domain of coronavirus.

To enable an efficient bacterial cell surface display with effective protein expression and cell surface loading ability via autotransporter for potential vaccine development applications, the inner membrane protein translocation efficiency was investigated via a trial-and-error strategy by replacing the original unusual long signal peptide of E. coli Ag43 with 11 different signal peptides. The receptor-binding domain (RBD) of coronavirus was used as a neutral display substrate to optimize the expression conditions, and the results showed that signal peptides from PelB, OmpC, OmpF, and PhoA protein enhance the bacterial cell surface display efficiency of RBD. In addition, the temperature has also a significant effect on the autodisplay efficiency of RBD. Our data provide further technical basis for the biotechnological application of Ag43 as a bacterial surface display carrier system and further potential application in vaccine development.

Characterization of the interaction between the Sec61 translocon complex and ppαF using optical tweezers.

The Sec61 translocon allows the translocation of secretory preproteins from the cytosol to the endoplasmic reticulum lumen during polypeptide biosynthesis. These proteins possess an N-terminal signal peptide (SP) which docks at the translocon. SP mutations can abolish translocation and cause diseases, suggesting an essential role for this SP/Sec61 interaction. However, a detailed biophysical characterization of this binding is still missing. Here, optical tweezers force spectroscopy was used to characterize the kinetic parameters of the dissociation process between Sec61 and the SP of prepro-alpha-factor. The unbinding parameters including off-rate constant and distance to the transition state were obtained by fitting rupture force data to Dudko-Hummer-Szabo models. Interestingly, the translocation inhibitor mycolactone increases the off-rate and accelerates the SP/Sec61 dissociation, while also weakening the interaction. Whereas the translocation deficient mutant containing a single point mutation in the SP abolished the specificity of the SP/Sec61 binding, resulting in an unstable interaction. In conclusion, we characterize quantitatively the dissociation process between the signal peptide and the translocon, and how the unbinding parameters are modified by a translocation inhibitor.

A new mRNA structure prediction based approach to identifying improved signal peptides for bone morphogenetic protein 2.

BACKGROUND: Signal peptide (SP) engineering has proven able to improve production of many proteins yet is a laborious process that still relies on trial and error. mRNA structure around the translational start site is important in translation initiation and has rarely been considered in this context, with recent improvements in in silico mRNA structure potentially rendering it a useful predictive tool for SP selection. Here we attempt to create a method to systematically screen candidate signal peptide sequences in silico based on both their nucleotide and amino acid sequences. Several recently released computational tools were used to predict signal peptide activity (SignalP), localization target (DeepLoc) and predicted mRNA structure (MXFold2). The method was tested with Bone Morphogenetic Protein 2 (BMP2), an osteogenic growth factor used clinically for bone regeneration. It was hoped more effective BMP2 SPs could improve BMP2-based gene therapies and reduce the cost of recombinant BMP2 production. RESULTS: Amino acid sequence analysis indicated 2,611 SPs from the TGF-ß superfamily were predicted to function when attached to BMP2. mRNA structure prediction indicated structures at the translational start site were likely highly variable. The five sequences with the most accessible translational start sites, a codon optimized BMP2 SP variant and the well-established hIL2 SP sequence were taken forward to in vitro testing. The top five candidates showed non-significant improvements in BMP2 secretion in HEK293T cells. All showed reductions in secretion versus the native sequence in C2C12 cells, with several showing large and significant decreases. None of the tested sequences were able to increase alkaline phosphatase activity above background in C2C12s. The codon optimized control sequence and hIL2 SP showed reasonable activity in HEK293T but very poor activity in C2C12. CONCLUSIONS: These results support the use of peptide sequence based in silico tools for basic predictions around signal peptide activity in a synthetic biology context. However, mRNA structure prediction requires improvement before it can produce reliable predictions for this application. The poor activity of the codon optimized BMP2 SP variant in C2C12 emphasizes the importance of codon choice, mRNA structure, and cellular context for SP activity.

Efficient signal sequence of mRNA vaccines enhances the antigen expression to expand the immune protection against viral infection.

The signal sequence played a crucial role in the efficacy of mRNA vaccines against virus pandemic by influencing antigen translation. However, limited research had been conducted to compare and analyze the specific mechanisms involved. In this study, a novel approach was introduced by substituting the signal sequence of the mRNA antigen to enhance its immune response. Computational simulations demonstrated that various signal peptides differed in their binding capacities with the signal recognition particle (SRP) 54 M subunit, which positively correlated with antigen translation efficiency. Our data revealed that the signal sequences of tPA and IL-6-modified receptor binding domain (RBD) mRNA vaccines sequentially led to higher antigen expression and elicited more robust humoral and cellular immune protection against the SARS-CoV-2 compared to the original signal sequence. By highlighting the importance of the signal sequence, this research provided a foundational and safe approach for ongoing modifications in signal sequence-antigen design, aiming to optimize the efficacy of mRNA vaccines.

Bacillus subtilis: current and future modification strategies as a protein secreting factory.

Bacillus subtilis is regarded as a promising microbial expression system in bioengineering due to its high stress resistance, nontoxic, low codon preference and grow fast. The strain has a relatively efficient expression system, as it has at least three protein secretion pathways and abundant molecular chaperones, which guarantee its expression ability and compatibility. Currently, many proteins are expressed in Bacillus subtilis, and their application prospects are broad. Although Bacillus subtilis has great advantages compared with other prokaryotes related to protein expression and secretion, it still faces deficiencies, such as low wild-type expression, low product activity, and easy gene loss, which limit its large-scale application. Over the years, many researchers have achieved abundant results in the modification of Bacillus subtilis expression systems, especially the optimization of promoters, expression vectors, signal peptides, transport pathways and molecular chaperones. An optimal vector with a suitable promoter strength and other regulatory elements could increase protein synthesis and secretion, increasing industrial profits. This review highlights the research status of optimization strategies related to the expression system of Bacillus subtilis. Moreover, research progress on its application as a food-grade expression system is also presented, along with some future modification and application directions.

Screening signal peptidase based on split-GFP assembly technology to promote the secretion of alkaline protease AprE in Bacillus amyloliquefaciens.

Improving the ability of bacteria to secrete protein is essential for large-scale production of food enzymes. However, due to the lack of effective tracking technology for target proteins, the optimization of the secretory system is facing many problems. In this study, we utilized the split-GFP system to achieve self-assembly into mature GFP in Bacillus amyloliquefaciens and successfully tracked the alkaline protease AprE. The split-GFP system was employed to assess the signal peptidases, a crucial component in the secretory system, and signal peptidase sipA was identified as playing a role in the secretion of AprE. Deletion of sipA resulted in a higher accumulation of the precursor protein of AprE compared to other signal peptidase deletion strains. To explore the mechanism of signal peptidase on signal peptide, molecular docking and calculation of free energy were performed. The action strength of the signal peptidase is determined by its binding affinity with the tripeptides at the C-terminal of the signal peptide. The functions of signal peptides YdbK and NucB rely on sipA, and overexpression of sipA by integrating it into genome of B. amyloliquefaciens increased the activity of extracellular AprE by 19.9 %. These findings provide insights into enhancing the secretion efficiency of chassis strains.

High-activity recombinant human carboxypeptidase B expression in Pichia pastoris through rational protein engineering and enhancing secretion from the Golgi apparatus to the plasma membrane.

Human carboxypeptidase B1 (hCPB1) is vital for recombinant insulin production, holding substantial value in the pharmaceutical industry. Current challenges include limited hCPB1 enzyme activity. In this study, recombinant hCPB1 efficient expression in Pichia pastoris was achieved. To enhance hCPB1 secretion, we conducted signal peptides screening and deleted the Vps10 sortilin domain, reducing vacuolar mis-sorting. Overexpression of Sec4p increased the fusion of secretory vesicles with the plasma membrane and improved hCPB1 secretion by 20%. Rational protein engineering generated twenty-two single-mutation mutants and identified the A178L mutation resulted in a 30% increase in hCPB1 specific activity. However, all combinational mutations that increased specific activities decreased protein expression levels. Therefore, computer-aided global protein design with PROSS was employed for the aim of improving specific activities and preserving good protein expression. Among the six designed mutants, hCPB1-P6 showed a remarkable 114% increase in the catalytic rate constant (kcat), a 137% decrease in the Michaelis constant (Km), and a 490% increase in catalytic efficiency. Most mutations occurred on the surface of hCPB1-P6, with eight sites mutated to proline. In a 5 L fermenter, hCPB1-P6 was produced by the secretion-enhanced P. pastoris chassis to 199.6 ± 20 mg L-1 with a specific activity of 96 ± 0.32 U mg-1, resulting in a total enzyme activity of 19137 ± 1131 U L-1, demonstrating significant potential for industrial applications.

Designing a Secretory form of RTX-A as an Anticancer Toxin: An In Silico Approach.

BACKGROUND: Cancer is a leading cause of death and a significant public health issue worldwide. Standard treatment methods such as chemotherapy, radiotherapy, and surgery are only sometimes effective. Therefore, new therapeutic approaches are needed for cancer treatment. Sea anemone actinoporins are pore-forming toxins (PFTs) with membranolytic activities. RTX-A is a type of PFT that interacts with membrane phospholipids, resulting in pore formation. The synthesis of recombinant proteins in a secretory form has several advantages, including protein solubility and easy purification. In this study, we aimed to discover suitable signal peptides for producing RTX-A in Bacillus subtilis in a secretory form. METHODS: Signal peptides were selected from the Signal Peptide Web Server. The probability and secretion pathways of the selected signal peptides were evaluated using the SignalP server. ProtParam and Protein-sol were used to predict the physico-chemical properties and solubility. AlgPred was used to predict the allergenicity of RTX-A linked to suitable signal peptides. Non-allergenic, stable, and soluble signal peptides fused to proteins were chosen, and their secondary and tertiary structures were predicted using GOR IV and I-TASSER, respectively. The PROCHECK server performed the validation of 3D structures. RESULTS: According to bioinformatics analysis, the fusion forms of OSMY_ECOLI and MALE_ECOLI linked to RTX-A were identified as suitable signal peptides. The final proteins with signal peptides were stable, soluble, and non-allergenic for the human body. Moreover, they had appropriate secondary and tertiary structures. CONCLUSION: The signal above peptides appears ideal for rationalizing secretory and soluble RTX-A. Therefore, the signal peptides found in this study should be further investigated through experimental researches and patents.

Phosphatidylglycerol Acts as a Switch for Cholesterol-Dependent Membrane Binding of ApoE Signal Peptide.

The apolipoprotein E (ApoE) signal peptide is a short stretch of N-terminal amino acids that direct the ApoE protein to the endoplasmic reticulum after synthesis. Previous studies have shown that this peptide can bind to lipid membranes in a cholesterol-dependent manner; however, the mechanism of this interaction is yet to be clarified. In this study, we aimed to investigate how the composition of neighboring lipids affects the membrane-binding of the ApoE signal peptide. We found that a negatively charged lipid, such as phosphatidylglycerol, can act as a switch that reduces the binding efficiency of the peptide to cholesterol-rich membranes. Interestingly, phosphatidylethanolamine does not activate the cholesterol-dependent binding of the ApoE signal peptide yet acts synergistically to enhance the cholesterol sensitivity in phosphatidylglycerol-containing membranes. To the best of our knowledge, this is the first report of modulation of the affinity of a peptide for a membrane by a neighboring lipid rather than by the lipid-binding domain of the peptide. Our findings revealed a novel role of lipid diversity in modulating the membrane binding of the ApoE signal peptide and its potential implications in the unidirectional trafficking of a newly synthesized protein from the ribosomes to the endoplasmic reticulum.

Optimizing Pichia pastoris protein secretion: Role of N-linked glycosylation on the α-mating factor secretion signal leader.

The methylotrophic yeast, Pichia pastoris (P. pastoris; syn. Komagataella spp.), known for its ability to grow to high cell densities, its strong and tightly regulated promoters, and mammalian liked secretion pathway, has been widely used as a robust system to secrete heterologous proteins. The α-mating factor (MF) secretion signal leader from Saccharomyces cerevisiae (S. cerevisiae) is currently the most successfully used secretion signal sequence in the P. pastoris system. In this study, the secretion efficiency mediated by the α-MF secretion signal leaders from Komagataella pastoris (K. pastoris) and Komagataella phaffii (K. phaffii) was assessed using Enhanced Green Fluorescent Protein (EGFP) as a reporter. The results indicated that the secretion efficiency associated with the α-MF secretion signal leaders from K. pastoris and K. phaffii was notably lower in comparison to the α-MF secretion signal leader from S. cerevisiae. Further research indicated that N-linked glycosylation of the α-MF secretion signal leader enhanced the secretion of EGFP. Disruption of calnexin impaired the secretion of EGFP mediated by the N-linked glycosylated α-MF secretion signal leader, without affecting EGFP secretion mediated by the non-N-linked glycosylation α-MF secretion signal leader. The N-linked glycosylated of the α-MF secretion signal leader reduced the unfolded protein response (UPR) in the endoplasmic reticulum (ER). The enhancement of EGFP secretion by the N-linked glycosylated α-MF secretion signal leader might be achieved through the acceleration of proper folding of glycoproteins by the molecular chaperone calnexin. This study enhances the understanding of protein secretion in P. pastoris, specifically highlighting the influence of N-linked glycosylation on secretion efficiency, and could have implications for the production of recombinant proteins in bioengineering and biotechnological applications in P. pastoris.

DeepLoc 2.1: multi-label membrane protein type prediction using protein language models.

DeepLoc 2.0 is a popular web server for the prediction of protein subcellular localization and sorting signals. Here, we introduce DeepLoc 2.1, which additionally classifies the input proteins into the membrane protein types Transmembrane, Peripheral, Lipid-anchored and Soluble. Leveraging pre-trained transformer-based protein language models, the server utilizes a three-stage architecture for sequence-based, multi-label predictions. Comparative evaluations with other established tools on a test set of 4933 eukaryotic protein sequences, constructed following stringent homology partitioning, demonstrate state-of-the-art performance. Notably, DeepLoc 2.1 outperforms existing models, with the larger ProtT5 model exhibiting a marginal advantage over the ESM-1B model. The web server is available at https://services.healthtech.dtu.dk/services/DeepLoc-2.1.