Antibody structures comprising helical framework loops

Abstract

Provided herein are antibody structures comprising at least one helical framework loop, related compositions and methods, such methods include methods of producing and using the antibody structures comprising at least one helical framework loop.

Description

[0001] ANTIBODY STRUCTURES COMPRISING HELICAL FRAMEWORK LOOPS RELATED APPLICATIONS This application claims the benefit under 35 U.S.C. § 119(e) to U.S. Provisional Application No.63 / 684,694, entitled “ANTIBODY STRUCTURES COMPRISING HELICAL FRAMEWORK LOOPS”, filed on August 19, 2024, and U.S. Provisional Application No.63 / 741,260, entitled “ANTIBODY STRUCTURES COMPRISING HELICAL FRAMEWORK LOOPS”, filed on January 2, 2025; the entire contents of each of which are incorporated herein by reference. FIELD OF THE INVENTION The invention relates at least in part to antibody structures comprising at least one helical framework loop, related compositions and methods, such methods include methods of producing the antibody structures comprising at least one helical framework loop as provided herein. SUMMARY OF THE INVENTION Provided herein are polypeptides or antibody structures with at least one framework loop comprising a conformation that is restricted to a rigid helical structure. The introduction of such a rigid structure in at least one framework loop can enable the use of protein design tools, which are largely ineffective in designing loops. Provided herein, in one aspect, is a polypeptide comprising at least one helical framework loop, wherein the at least one helical framework loop comprises a rigid structure. In one embodiment of any one of the compositions or methods provided herein, the rigid structure is an alpha helix secondary structure. In one embodiment of any one of the compositions or methods provided herein, the polypeptide or antibody structure can bind a target. In one embodiment of any one of the compositions or methods provided herein, the target is IL7Rα, IL23α, TL1A or TNFα. In one embodiment of any one of the compositions or methods provided herein, the antibody structure has a binding affinity of at least 10-6M or 10-7M. In one embodiment of any one of the compositions or methods provided herein, the antibody structure has a binding affinity of no more than 10-8M or no more than 10-9M. In one embodiment of any one of the compositions or methods provided herein, the antibody structure has a binding affinity of between 10-6M and 10-9M, 10-7M and 10-9M or 10-8M and 10-9M. In one embodiment of any one of the compositions or methods provided herein, the antibody structure has a binding affinity of 10-6M, 10-7M, 10-8M or 10-9M. In one embodiment of any one of the compositions or methods provided herein, the antibody structure can be stable or resistant to degradation at a gastric pH, such as at a pH of 2. In one embodiment of any one of the compositions or methods provided herein, the antibody structure can be stable or resistant to degradation in simulated gastric fluid, such as the simulated gastric fluid of Table 2. In one embodiment of any one of the compositions or methods provided herein, the antibody structure can be stable or resistant to degradation in simulated gastric fluid, such as the simulated gastric fluid of Table 2, for at least 1 hour, 2 hours, 3 hours or 4 hours. In one embodiment of any one of the compositions or methods provided herein, the antibody structure can be stable or resistant to degradation at an intestinal pH, such as at a pH of 6.5. In one embodiment of any one of the compositions or methods provided herein, the antibody structure can be stable or resistant to degradation in simulated intestinal fluid, such as the simulated intestinal fluid of Table 3. In one embodiment of any one of the compositions or methods provided herein, the antibody structure can be stable or resistant to degradation in simulated intestinal fluid, such as the simulated intestinal fluid of Table 3, for at least 1 hour, 2 hours, 3 hours or 4 hours. In one embodiment of any one of the compositions or methods provided herein, the antibody structure can be stable or resistant to degradation in the presence of a protease. In one embodiment of any one of the compositions or methods provided herein, the antibody structure is stable or resistant to degradation in the presence of any one or more of the proteases provided herein. In one embodiment of any one of the compositions or methods provided herein, the antibody structure is stable or resistant to degradation in the presence of trypsin and / or chymotrypsin. In one embodiment of any one of the compositions or methods provided herein, the antibody structure is stable or resistant to degradation in the presence of pepsin. In one embodiment of any one of the compositions or methods provided herein, the antibody structure is thermostable. In one embodiment of any one of the compositions or methods provided herein, the antibody structure is stable or resistant to degradation at a temperature of at least or equal to 37ºC. In one embodiment of any one of the compositions or methods provided herein, the antibody structure is stable or resistant to degradation at a temperature of at least or equal to any one of the temperatures provided herein. In one embodiment of any one of the compositions or methods provided herein, the antibody structure is stable or resistant to degradation at a temperature of at least or equal to 70ºC, 75ºC, 80ºC, 85ºC, 90ºC or 95ºC. In one embodiment of any one of the compositions or methods provided herein, the antibody structures of a composition in which they are comprised are monodispersed and / or do not aggregate in the composition. In one embodiment of any one of the compositions or methods provided herein, the antibody structures provided herein comprise at least two helical framework loops, wherein each of the least two helical framework loops comprise a rigid structure. In one embodiment of any one of the compositions or methods provided herein, the antibody structures comprise at least three helical framework loops, wherein each of the at least three helical framework loops comprise a rigid structure. In one embodiment of any one of the compositions or methods provided herein, the antibody structures comprise four helical framework loops, wherein each of the four helical framework loops comprise a rigid structure. In one embodiment of any one of the compositions or methods provided herein, the antibody structures does not comprise a rigid structure, such as an alpha helix secondary structure, in a CDR of the antibody structure. In one embodiment of any one of the compositions or methods provided herein, the antibody structures further comprise an antibody scaffold. In one embodiment of any one of the compositions or methods provided herein, the antibody scaffold comprises a single-domain antibody (sdAb), single-chain variable fragment (scFv), Fab’, fragment antigen binding (Fab), F(ab’)2or a full-length antibody without the framework loop(s) replaced by the helical framework loop(s) of the antibody structure. In one embodiment of any one of the compositions or methods provided herein, the at least one helical framework loop replaces the loop of framework region 1 (FR1). In one embodiment of any one of the compositions or methods provided herein, the at least one helical framework loop replaces the loop of FR2. In one embodiment of any one of the compositions or methods provided herein, the at least one helical framework loop replaces a loop of FR3. In one embodiment of any one of the compositions or methods provided herein, the at least one helical framework loop replaces the first loop of FR3. In one embodiment of any one of the compositions or methods provided herein, the at least one helical framework loop replaces the second loop of FR3. In one embodiment of any one of the compositions or methods provided herein, the at least one helical framework loop replaces the third loop of FR3. In one embodiment of any one of the compositions or methods provided herein, the at least one helical framework loop replaces the loop of FR1 and / or the third loop of FR3. In one embodiment of any one of the compositions or methods provided herein, the at least one helical framework loop replaces the loop of FR2 and / or the third loop of FR3. In one embodiment of any one of the compositions or methods provided herein, the at least one helical framework loop replaces the loop of FR1 and / or the first loop of FR3. In one embodiment of any one of the compositions or methods provided herein, the at least one helical framework loop replaces the loop of FR2 and / or the first loop of FR3. In one embodiment of any one of the compositions or methods provided herein, the at least one helical framework loop replaces the loop of FR1 and / or the loop of FR2. In one embodiment of any one of the compositions or methods provided herein, the at least one helical framework loop replaces the loop of FR1 and / or the loop of FR2 and / or the third loop of FR3. In one embodiment of any one of the compositions or methods provided herein, the at least one helical framework loop replaces the loop of FR1 and / or the loop of FR2 and / or the first loop of FR3. In one embodiment of any one of the compositions or methods provided herein, the at least one helical framework loop replaces the loop of FR1 and / or the loop of FR2 and / or the first loop of FR3 and / or the third loop of FR3. In one embodiment of any one of the compositions or methods provided herein, each helical framework loop is 5-20 residues in length. In one embodiment of any one of the compositions or methods provided herein, each helical framework loop is 5-19 residues in length. In one embodiment of any one of the compositions or methods provided herein, each helical framework loop is 5-18 residues in length. In one embodiment of any one of the compositions or methods provided herein, each helical framework loop is 5-17 residues in length. In one embodiment of any one of the compositions or methods provided herein, each helical framework loop is 5-16 residues in length. In one embodiment of any one of the compositions or methods provided herein, each helical framework loop is 5-15 residues in length. In one embodiment of any one of the compositions or methods provided herein, each helical framework loop is 5-14 residues in length. In one embodiment of any one of the compositions or methods provided herein, each helical framework loop is 5-13 residues in length. In one embodiment of any one of the compositions or methods provided herein, each helical framework loop is 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19 or 20 residues in length. In one embodiment of any one of the compositions or methods provided herein, each helical framework loop used to replace the loop of FR1 is 5-20 residues in length. In one embodiment of any one of the compositions or methods provided herein, each helical framework loop used to replace the loop of FR1 is 5-13 residues in length. In one embodiment of any one of the compositions or methods provided herein, each helical framework loop used to replace the third loop of FR3 is 5-20 residues in length. In one embodiment of any one of the compositions or methods provided herein, each helical framework loop used to replace the third loop of FR3 is 5-19 residues in length. In one embodiment of any one of the compositions or methods provided herein, each helical framework loop used to replace the loop of FR2 is 5-18 residues in length. In one embodiment of any one of the compositions or methods provided herein, each helical framework loop used to replace the loop of FR2 is 5-14 residues in length. In one embodiment of any one of the compositions or methods provided herein, each helical framework loop used to replace the first loop of FR3 is 5-18 residues in length. In one embodiment of any one of the compositions or methods provided herein, each helical framework loop used to replace the first loop of FR3 is 5-14 residues in length. In one embodiment of any one of the compositions or methods provided herein, the antibody structure further comprises at least one linker at the N-terminus and / or C-terminus of at least one helical framework loop. In one embodiment of any one of the compositions or methods provided herein, the antibody structure further comprises at least one linker at the N-terminus and / or C- terminus of each helical framework loop. In one embodiment of any one of the compositions or methods provided herein, the linker does not form an alpha helix or beta sheet secondary structure. In one embodiment of any one of the compositions or methods provided herein, the linker at an N-terminus is 1-8 residues in length. In one embodiment of any one of the compositions or methods provided herein, the linker at an N-terminus is 1-5 residues in length. In one embodiment of any one of the compositions or methods provided herein, the linker at an N-terminus is 1, 2, 3, 4, 5, 6, 7, or 8 residues in length. In one embodiment of any one of the compositions or methods provided herein, the linker at a C-terminus is 1-8 residues in length. In one embodiment of any one of the compositions or methods provided herein, the linker at a C-terminus is 1-5 residues in length. In one embodiment of any one of the compositions or methods provided herein, the linker at a C-terminus is 1, 2, 3, 4, 5, 6, 7, or 8 residues in length. In one embodiment of any one of the compositions or methods provided herein, the antibody structure does not comprise a linker at the N-terminus and / or C-terminus of at least one helical framework loop. In one embodiment of any one of the compositions or methods provided herein, the antibody structure does not comprise a linker at the N-terminus and / or C-terminus of each helical framework loop. In one embodiment of any one of the compositions or methods provided herein, the antibody structures have or are selected or designed to have specific features, such as any one or or more, or all, of the features provided herein. Such features, include, but are not limited to, being stable or resistant to degradation in an intestinal environment or a gastric environment and / or being stable or resistant to degradation by one or more proteases and / or thermostability, as provided herein. In one embodiment of any one of the compositions or methods provided herein, the alpha helices comprise the sequence of any one or more or all of the alpha helices provided herein, such as those in the Examples, such as those shown in bold. In one aspect are polypeptides comprising any one of the antibody structures provided herein. In one aspect are compositions comprising any of the polypeptides (or antibody structures) provided herein. In one embodiment of any one of the compositions provided herein, a composition further comprises a pharmaceutically acceptable carrier. In one aspect, a method of administering any one of the antibody structures or any one of the polypeptides or any one of the compositions provided herein to a subject is provided. In one embodiment, the administering is done by oral delivery. In one aspect, a method for producing any one of the antibody structures provided herein is provided. In one aspect, the method is any one of the methods provided in the Examples. In one embodiment of any one of the methods provided herein, a method for producing an antibody structure comprising at least one helical framework loop that binds a target, comprises a) computationally generating a backbone (or obtaining such a backbone) comprising at least one alpha helix (such as any one of the alpha helices or sets of alpha helices provided herein) and an antibody scaffold structure. In another aspect, the method further comprises b) generating an amino acid sequence for the backbone of a), and c) optionally, evaluating the produced polypeptide using a protein structure prediction model. In one embodiment of any one of the methods provided herein, a) comprises using a diffusion model with fold conditioning to specify at least one fold. In one embodiment of any one of the methods provided herein, a) comprises providing a target structure and target residues on the target structure using a diffusion model with fold conditioning. In one embodiment of any one of the methods provided herein, the fold conditioning comprises specifying secondary structural element(s). In one embodiment of any one of the methods provided herein, the secondary structural element(s) comprise the at least one alpha helix and, optionally, one or more additional structural elements (e.g., beta sheet, loop, etc.) of an antibody scaffold. In one embodiment of any one of the methods provided herein, the fold is specified by providing each structural element, its length in number of residues, and its contacts with other structural element(s). In one embodiment of any one of the methods provided herein, i) a backbone with at least one helical framework loop is generated with a target structure and target residues on that structure (e.g., the epitope) using a diffusion model with fold conditioning to specify at least one fold. In one embodiment of any one of the methods provided herein, i) a backbone is generated without providing a target using a diffusion model with fold conditioning to specify at least one fold, and ii) a backbone is generated to bind to a target using a diffusion model with fold conditioning as well as a target structure and target residues on that structure. In one embodiment of the foregoing, at least one helix of a helical fold is grafted onto the backbone of i). In one embodiment of any one of the methods provided herein, a) comprises i) generating a backbone comprising an antibody scaffold structure using a diffusion model with fold conditioning, ii) generating a backbone comprising at least two alpha helices, such as one in the FR1 loop and the other in the third loop of FR3, that bind to a target using a diffusion model with fold conditioning as well as a target structure and target residues on the target structure, and iii) grafting two of the helices of ii) onto the backbone of i). In one embodiment of the foregoing, the fold conditioning comprises specifying secondary structural element(s). In one embodiment of any one of the methods provided herein, the secondary structural elements for i) comprise one or more structural elements of an antibody scaffold without providing a target. In one embodiment of any one of the foregoing, b) occurs before or after iii). In one embodiment of any one of the methods provided herein, i) a backbone is generated without providing a target using a diffusion model with fold conditioning to specify at least one fold, and ii) a backbone with at least one helix in a helical fold is generated to bind to a target, using a diffusion model with fold conditioning as well as a target structure and target residues on that structure. In one embodiment of the foregoing, the at least one helix in the helical fold is then used to dock the antibody structure without grafting. In one embodiment of any one of the methods provided herein, b) comprises using a message-passing neural nework model. In one embodiment of any one of the methods provided herein where grafting is performed, the sequences can be generated before or after the helix or helices are grafted onto the antibody structure. The sequences for a target and the antibody scaffold can be provided to the model along with a backbone structure of a complex (which includes the target and antibody with helical framework loop(s)), and the sequence is generated for the helical framework loops in this context. If beneficial to improve the predicted accuracy and confidence in the design, some of the amino acids of the antibody scaffold sequence may be modified to better support the helical framework loops. Any one of the relevant methods provided herein can include such a step. In one embodiment of the foregoing, the amino acids neighboring the helical framework loops are modified. In one embodiment of any one of the methods provided herein, a diffusion model may also be used in a ‘partial diffusion’ mode to modify a designed antibody structure with helical framework loops. In one embodiment, such a method comprises adding noise to the backbone coordinates, then using a diffusion model to generate a new backbone from the ‘noisy’ backbone. Such a method may further comprises generating a new sequence for this new backbone. In one embodiment of any one of the methods provided herein, the method comprises a step of evaluating an antibody structure using a protein structure prediction model. Such a model can provide a predicted structure given the sequence of a design, which can be aligned to the design structure model to assess accuracy and / or provide a measure of confidence in the prediction. In one embodiment of any one of the methods provided herein, the method further comprises redesigning amino acids of the antibody scaffold sequence to better support the helical framework region(s). In one embodiment of any one of the methods of production provided herein, any one or more or all of the steps of the methods of the Examples may be included. In one embodiment of any one of the methods of production provided herein, any one of the steps can comprise any one or more or all of the relevant steps of the Examples. In one embodiment of any one of the methods provided herein, amino acid sequences are generated or selected that provide the antibody structures with specific features, such as any one or more or all of the features provided herein. Such features, include but are not limited to, being resistant to an acidic environment, such as to the highly acidic pH of the stomach (e.g., a pH of 1, 1.5, 2, 2.5, 3 or 3.5), thermostability, being resistant to one or more proteases and / or does not aggregate in a composition as provided herein. In one embodiment of any one of the methods provided herein, the method comprises or further comprises assessing, modeling or selecting for resistance to degradation in acidic conditions, such as at a pH of the stomach, thermostability, shelf stability, oral deliverability and / or resistance to protease or proteolytic degradation. In one embodiment of any one of the methods provided herein, the method comprises or further comprises assessing, modeling or selecting for resistance to degradation at an intestinal pH, such as a pH of 6.5. In one embodiment of any one of the methods provided herein, the method comprises or further comprises assessing, modeling or selecting for resistance to degradation in simulated intestinal fluid, such as that of Table 3. In one embodiment of any one of the methods provided herein, the method comprises or further comprises assessing, modeling or selecting for resistance to degradation at a gastric pH, such as a pH of 2. In one embodiment of any one of the methods provided herein, the method comprises or further comprises assessing, modeling or selecting for resistance to degradation in simulated gastric fluid, such as that of Table 2. In one embodiment of any one of the methods provided herein, the method comprises or further comprises assessing, modeling or selecting for resistance to degradation at 37ºC or any one of the temperatures provided herein. In one embodiment of any one of the methods provided herein, the method comprises or further comprises assessing, modeling or selecting for resistance to degradation by one or more proteases. In one embodiment of any one of the methods provided herein, the one or more proteases comprises trypsin and / or chymotrypsin. In one embodiment of any one of the methods provided herein, the one or more proteases comprises pepsin. In one embodiment of any one of the methods provided herein, the polypeptides can be orally delivered and / or shelf stable (such as not requiring cold storage, such as not requiring refrigeration). Any one of the compositions provided herein may be orally delivered and / or shelf stable. Thus, the polypeptides or compositions provided herein may be stored at room temperature or above. Any one of the methods provided herein can include a step of generating, selecting or designing a polypeptide with the foregoing features alone or in combination.

[0002] BRIEF DESCRIPTION OF THE FIGURES FIG.1 provides a schematic depicting (a) a standard nanobody fold and (b) a nanobody fold with helical framework loops in loop 1 of FR1 and loop 3 of FR3. FIG.2 provides an example nanobody (VHH) fold with helical framework loops in loop 1 of FR1 and loop 3 of FR3. FIG.3 provides an example nanobody (VHH) fold with helical framework loops in loop 1 of FR1 and loop 3 of FR3 (alternate view). FIG.4 illustrates a nanobody with two helical framework loops designed to bind to IL7Rα. The design model is aligned to the predicted structure. FIG.5 illustrates a nanobody with two helical framework loops designed to bind to IL7Rα. Shown is the interface detail of the predicted structure. FIGS.6 and 7 provide a BLI binding curve and nano DSF melting temperature curve, respectively, for an exemplary designed single domain antibody with helical framework loops that target TNFα. FIGS.8-11 provide BLI binding curves and nano DSF melting temperature curves for exemplary designed single domain antibodies with helical framework loops that target IL7Rα. FIGS.12-31 provide BLI binding curves and nano DSF melting temperature curves for exemplary designed single domain antibodies with helical framework loops that target IL23α. FIGS.32-37 provide results from simulated gastric and simulated intestinal fluid incubation experiments using designed single domain antibodies. The results show an estimated survival time of at least 4 hours in simulated gastric fluid (SGF) and simulated intestinal fluid (SIF). FIGS.38 and 39 provide size exclusion chromatography traces of example antibody samples, which show the samples were monodisperse and do not aggregate. FIGS.40-45 provide predicted structures of designed single domain antibodies with helical framework loops. DETAILED DESCRIPTION OF THE INVENTION Before describing the present invention in detail, it is to be understood that this invention is not limited to particularly exemplified materials or process parameters as such may, of course, vary. It is also to be understood that the terminology used herein is for the purpose of describing particular embodiments of the invention only, and is not intended to be limiting of the use of alternative terminology to describe the present invention. All publications, patents and patent applications cited herein, whether supra or infra, are hereby incorporated by reference in their entirety for all purposes. Such incorporation by reference is not intended to be an admission that any of the incorporated publications, patents and patent applications cited herein constitute prior art. As used in this specification and the appended claims, the singular forms "a," "an" and "the" include plural referents unless the content clearly dictates otherwise. For example, reference to "a molecule" includes a mixture of two or more such molecules or a plurality of such molecules, and the like. As used herein, the term “comprise” or variations thereof such as “comprises” or “comprising” are to be read to indicate the inclusion of any recited integer (e.g. a feature, element, characteristic, property, method / process step or limitation) or group of integers (e.g. features, elements, characteristics, properties, method / process steps or limitations) but not the exclusion of any other integer or group of integers. Thus, as used herein, the term “comprising” is inclusive and does not exclude additional, unrecited integers or method / process steps. In embodiments of any of the compositions and methods provided herein, “comprising” may be replaced with “consisting essentially of” or “consisting of”. The phrase “consisting essentially of” is used herein to require the specified integer(s) or steps as well as those which do not materially affect the character or function of the claimed invention. As used herein, the term “consisting” is used to indicate the presence of the recited integer (e.g. a feature, element, characteristic, property, method / process step or limitation) or group of integers (e.g. features, elements, characteristics, properties, method / process steps or limitations) alone. Introduction Generally, the antibody drug development process consists of three broad stages: discovery, optimization, and development. Discovery refers to the process of finding an initial antibody sequence or sequences. Optimization involves altering this sequence to improve desired properties, for example binding affinity or stability. Development includes all steps after the final antibody sequence is fixed, in which formulation and manufacturing are established for the drug candidate, animal testing is used to assess safety and determine first-in-human dose, and clinical trials are conducted. The discovery stage includes methods such as immunization, display technologies (e.g., yeast, phage), and B cell isolation, which may be used separately or in combination to discover drug candidates from very large (i.e., >108) libraries of potential sequences. Antibody libraries may consist of the native antibody repertoire from animals (including somatic hypermutation of antibodies recognizing a target) or synthetic libraries designed by scientists or generated through machine learning. Current discovery methods screen these libraries in order to identify antibody sequences that bind to a given target. These discovery methods provide very limited control over the epitope to which the antibody paratope binds. Furthermore, the experimental conditions associated with these methods may not accurately represent the target in its native context, leading to ineffective drugs in later stages of development. Candidates discovered through these methods do not achieve the full set of the desired drug properties, because their discovery is limited by traits specific to the selection methods (e.g., high surface expression on yeast for yeast display). As a result, candidate antibodies may have limited efficacy when moved to the native system and may not exhibit the properties required for manufacturing and formulation as an antibody drug. These candidate antibodies require significant optimization and development in order to produce functional drugs. The optimization stage involves the generation (either manually or via machine learning) and testing of further libraries consisting of variants of the candidate antibodies in order to identify high- value variants with desired drug properties. This includes high-throughput methods such as display technologies and low-throughput methods such as biolayer interferometry. Antibody optimization is typically an iterative process where select properties may be improved individually or in combination. The final selection of high-value variants requires balancing tradeoffs between multiple desired drug properties. The development stage consists of the remaining steps in the pipeline prior to the final drug production, including pre-clinical trials, clinical trials, and any additional development of the drug. Due to the difficulties with performing the above steps, alternative methodology for antibody development have been employed. For example, CDR grafting, which refers to the transfer of CDR loops from one antibody onto another through replacement of the CDR sequence, has been a part of the antibody engineering toolkit for decades. In addition, the grafting of peptides to replace one or multiple CDR loops has also been used in specific cases to enable antibody design (8, 10). In recent years, machine learning has been applied to protein structure prediction to achieve high-accuracy prediction (up to atomic-accuracy) for many proteins (1, 2). However, these methods are typically unable to accurately predict the structure of antibody CDR loops due to their high conformational diversity (3). Thus, antibody drug development has not fully benefited from such tools, and it has remained difficult to predict the structure and binding of antibodies. Structure prediction is a determinant of the relative difficulty and potential accuracy of protein design. Generally, high-affinity binding proteins (“binders”) with the ability to bind to a wide range of targets can be created (4). Such processes can be improved through the incorporation of machine learning methods including diffusion models and graph neural networks (5-7). The percentage of designed binders that achieve binding to desired targets can exceed 10% with up to picomolar affinity (5, 6). Such processes do not require iterative optimization but rather rely on the design of de novo proteins with binding interfaces that feature well-defined secondary structural element(s) (e.g., alpha helices and beta sheets). The inventor has surprisingly and unexpectedly discovered that the problems and limitations noted above can be overcome by practicing the invention disclosed herein. The methods and compositions provided offer solutions to the aforementioned obstacles to effective design of antibody structures. A key feature in the development of the antibody structures provided herein is the introduction of a helical backbone which restricts conformation(s) of framework loop region(s) to a rigid helical structure. Preferably, this rigid structure is conferred with an alpha helix. This allows the use of protein design tools. The antibody structures are each defined by a specific number of helical framework loops and an antibody scaffold. Thus, compositions comprising the antibody structures provided herein are provided as are methods of producing and using such antibody structures. The invention will now be described in more detail below. Compositions of Antibody Structures with Helical Framework Loop(s) Antibody framework loops built on helical backbones have highly predictable and controllable structure, allowing for precise sequence and structural manipulation. Helical framework loop(s) can have highly predictable and controllable structure and can allow for precise sequence and structural manipulation. Helical framework loops can allow for the generation of antibody structures with a variety of properties. These properties include, but are not limited to, target epitope binding, reduced off-target binding, binding affinity, aggregation, pharmacokinetic properties, etc. The design capabilities provided herein can also enable the avoidance of liabilities like glycosylation, deamidation, isomerization, cleavage, and hydrolysis sites, as well as T and B cell epitopes, in some embodiments. Still other properties include, but are not limited to, features such as being pH responsive, thermostable, etc. Further, antibody structures can be designed that avoid undesirable features, such as degradation in an acidic environment, such as the highly acidic pH of the stomach, and / or degradation by one or more proteases. The antibody structures provided herein comprise at least one helical framework and can bind a target in some embodiments. As used herein, a “target” is any molecule to which specific binding is desired. As used herein, “specific binding” refers to a molecule binding to a predetermined target with at least two-fold greater affinity than its affinity for binding to a non-specific target. In one embodiment, the antibody structure has a binding affinity of at least 10-6M or 10-7M. In one embodiment, the antibody structures have a binding affinity of at least 10-8M or 10-9M. In one embodiment of any one of the compositions or methods provided herein, the antibody structure has a binding affinity of between 10-6M and 10-9M, 10-7M and 10-9M or 10-8M and 10-9M. In one embodiment of any one of the compositions or methods provided herein, the antibody structure has a binding affinity of 10-6M, 10-7M, 10-8M or 10-9M or better. The binding affinity of the antibody structures can be determined by a variety of methods known in the art. For example, binding affinity can be determined as provided in the Examples. The binding affinity can also be determined, for example, by loading biotinylated target antigen onto streptavidin coated biosensor and measuring association and dissociation of His-tag purified antibody structure. Binding affinity can also be determined by Enzyme-Linked Immunosorbent Assay (ELISA), Surface Plasmon Resonance (SPR), Mass Photometry (MP), Cell-Based Fluorescent Assays and Chaotrope-Based Assays, in embodiments. The at least one helical framework loop of the antibody structures provided herein comprises an alpha helix secondary structure and is used in place of a typical antibody framework loop, in an embodiment. In another embodiment, the helical framework loops consist of backbones adopting an alpha helical secondary structure. The helical framework loop(s) provided herein can be used in place of an antibody FR1, FR2, or FR3 loop. In the antibody structures provided herein, there may be helical framework loops that replace one, two, three or four of the foregoing loops. In some embodiments, the helical framework loop replaces the FR1, the FR2 or a FR3 (e.g., first or third) loop. In some embodiments, the helical framework loop replaces the FR1 and / or the first FR3 loop. In some embodiments, the helical framework loop replaces the FR1 and / or the third FR3 loop. In some embodiments, the helical framework loop replaces the FR1 and / or the FR2 loop. In some embodiments, helical frameworks loops replace FR1, FR2 and FR3 (first or third) loops. In some embodiments, helical frameworks loops replace FR1, FR2 and FR3 (first and third) loops. The helical framework loop(s) can have lengths within specified ranges for the number of residues. The lower bound of these ranges is generally the minimum number of residues that can be both sufficient to design binding to a target and adopt an alpha helical secondary structure with preferably high predicted accuracy and confidence. Below the minimum number the helical framework loop can become unstructured and cannot be designed to bind to the target in some embodiments. The upper bound of these ranges is generally the maximum number of residues that can be accommodated by the antibody scaffold and adopt an alpha helical secondary structure with predicted accuracy and confidence. Above the maximum number the helical framework loop can introduce too much strain into the structure and may not stably fold in some embodiments. It should be noted that while some preferred ranges include helical framework loops that can be predicted with high accuracy and confidence and can be designed effectively to bind to a target, helical framework loops that are predicted with a lower accuracy and confidence can still be able to be designed to bind to a target in some embodiments. The helical framework loops, preferably, are of a length that preferably provides specific binding to a target and adopts an alpha helical secondary structure but that does not introduce too much strain into the structure such that stable folding is negatively impacted. For example, a helical framework loop can have a length of 5-20 residues, 5-19 residues, 5-18 residues, 5-17 residues, 5-16 residues, 5-15 residues, 5-14 residues, 5-13, residues, 5-12 residues, 5-11 residues, 5-10 residues, 5-9 residues, 5-8 residues, or 5-7 residues. As another example, a helical framework loop can have a length of 6-20 residues, 6-19 residues, 6-18 residues, 6-17 residues, 6-16 residues, 6-15 residues, 6- 14 residues, 6-13, residues, 6-12 residues, 6-11 residues, 6-10 residues, 6-9 residues, or 6-8 residues. As a further example, a helical framework loop can have a length of 7-20 residues, 7-19 residues, 7- 18 residues, 7-17 residues, 7-16 residues, 7-15 residues, 7-14 residues, 7-13, residues, 7-12 residues, 7-11 residues, 7-10 residues, or 7-9 residues. As still a further example, a helical framework loop can have a length of 8-20 residues, 8-19 residues, 8-18 residues, 8-17 residues, 8-16 residues, 8-15 residues, 8-14 residues, 8-13, residues, 8-12 residues, 8-11 residues, or 8-10 residues. As a further example, a helical framework loop can have a length of 9-20 residues, 9-19 residues, 9-18 residues, 9-17 residues, 9-16 residues, 9-15 residues, 9-14 residues, 9-13, residues, 9-12 residues, or 9-11 residues. As yet a further example, a helical framework loop can have a length of 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19 or 20 residues. Other lengths may be determined by the number of residues that is both sufficient to design binding to a target and that adopts an alpha helical secondary structure with predicted accuracy and confidence. Other lengths may be determined by the number of residues that can be accommodated by an antibody scaffold and that adopts an alpha helical secondary structure with high predicted accuracy and confidence. Generally, an antibody scaffold with the helical framework loop(s) make up the antibody structures as provided herein. As used herein, an antibody scaffold refers to a structure of an antibody or antibody-like molecule but without framework loop(s) that are replaced with the helical framework loop(s) as provided herein. In some embodiments, the antibody scaffold comprises framework loop(s) that are not replaced with a helical framework loop. In some embodiments, however, the antibody scaffold does not comprise any framework loops that are not replaced with a helical framework loop. In some embodiments, antibody scaffolds and sequences can come from known antibodies or newly discovered antibodies. The antibody scaffolds may be from a crystal structure or a predicted structure, in other embodiments. The antibody structures provided herein can comprise any one of the antibody scaffolds provided herein. Typical antibodies are generally glycoproteins comprising at least two heavy (H) chains and two light (L) chains inter-connected by disulfide bonds. Each heavy chain is comprised of a heavy chain variable region (abbreviated herein as HCVR or VH) and a heavy chain constant region. The heavy chain constant region is comprised of three domains, CH1, CH2 and CH3. Each light chain is comprised of a light chain variable region (abbreviated herein as LCVR or VL) and a light chain constant region. The light chain constant region is comprised of one domain, CL. The VH and VL regions can be further subdivided into regions of hypervariability, termed complementarity determining regions (CDRs), interspersed with regions that are more conserved, termed framework regions (FRs). Each VH and VL is composed of three CDRs and four FRs, arranged from amino- terminus to carboxy-terminus in the following order: FR1, CDR1, FR2, CDR2, FR3, CDR3, FR4. The variable regions of the heavy and light chains contain a binding domain that interacts with a target (an antigen). The constant regions of the antibodies may mediate the binding of the immunoglobulin to host tissues or factors, including various cells of the immune system (e.g., effector cells) and the first component (C1q) of the classical complement system. Antigen-binding fragments of a typical antibody refers to one or more portions of an antibody that retain the ability to bind specifically to a target (antigen). The antigen-binding function of an antibody can be performed by fragments of a full-length antibody. Examples of antigen-binding fragments include (i) a Fab fragment, a monovalent fragment consisting of the VL, VH, CLand CH1 domains; (ii) a F(ab′)2 fragment, a bivalent fragment comprising two Fab fragments linked by a disulfide bridge at the hinge region; (iii) a Fd fragment consisting of the VH and CH1 domains; (iv) a Fv fragment consisting of the VL and VH domains of a single arm of an antibody, (v) a dAb fragment (Ward et al., (1989) Nature 341:544-546), which consists of a VH domain., and (vi) a sdAb, single-domain antibody. Furthermore, although the two domains of the Fv fragment, V and VH, are coded for by separate genes, they can be joined, using recombinant methods, by a synthetic linker that enables them to be made as a single protein chain in which the VL and VH regions pair to form monovalent molecules (known as single chain Fv (scFv); see e.g., Bird et al. (1988) Science 242:423-426; and Huston et al. (1988) Proc. Natl. Acad. Sci. USA 85:5879-5883). Such single chain antibodies are also considered an example of antigen-binding fragment of a typical antibody. The antibody scaffolds of the compositions and methods provided herein can be any one of the foregoing structures but without framework loop(s) that are replaced with the helical framework loop(s) or without any framework loop(s) as provided herein. In some embodiments of any one of the compositions or methods provided herein, the antibody scaffold is a VHH (also known as nanobody, single-domain antibody) or scFv (antibodies with both VH and VL chains). A key feature in the development of the antibody structures provided herein is the introduction of a rigid structure, which restricts a conformation(s). Rigid structures as provided herein can have highly predictable and controllable structure and can allow for precise sequence and structural manipulation. As used herein, a “rigid structure” is any structure that can restrict a conformation and that can allow the use of protein design tools, such as those which otherwise may be largely ineffective in designing an antibody or antibody-like molecule. The rigid structure may be an alpha helix. “Alpha helix” refers to a sequence of amino acids that can be twisted into a coil (a helix). Generally, the alpha helix has a right-handed helix conformation in which backbone N−H groups hydrogen bond to the backbone C=O groups of the amino acid that is four residues earlier in the protein sequence, respectively. In one embodiment of any one of the compositions or methods provided herein, the alpha helix may have the sequence of any one of the alpha helices provided herein. The antibody structures provided herein can include linkers. The linkers can join the helical framework loop(s) to an antibody scaffold with unstructured segments that, preferably, do not adopt an alpha helix or beta sheet secondary structure and / or are distinct from the helical framework loop(s) and the beta sheets of the antibody scaffold that they join. These segments or “linkers” can be before and / or after a helical framework loop in the peptide chain (from N-terminus to C- terminus). However, these linkers may not be necessary in some cases (equivalent to being 0 residues in length). In some embodiments of any one of the compositions or methods provided herein, at least one linker for a helical framework loop at the N-terminus and / or C-terminus must be longer than 0 residues to avoid strain in the structure and / or to ensure stable folding. In one embodiment of any one of the methods provided herein, there are linkers before and / or after a helical FR1 loop and / or before and / or after a helical FR2 loop and / or before and / or after the first loop of FR3 and / or before and / or after the third loop of FR3 in the peptide chain (from N terminus to C terminus). In one embodiment of any one of the methods provided herein, there are linkers before and / or after a helical FR1 loop and / or before and / or after the third loop of FR3 in the peptide chain (from N terminus to C terminus). These linkers may not be necessary in some cases (equivalent to being 0 residues in length), but typically at least one linker for a helical framework loop must be longer than 0 residues to avoid strain in the structure and to ensure stable folding. Above a maximum residue length, the linkers can become a liability and result in instability of the structure and / or aggregation of the antibodies. Within preferred length ranges the resulting antibodies can be predicted with high accuracy and confidence and can be designed effectively to bind to a target; however, even if antibodies are less likely to be stable they may still function as designed in some embodiments. Amino acids for the linkers can comprise glycine, proline, and / or serine, but other amino acids are also possible as are other linkers. Preferably, however, the linker backbones do not adopt an alpha helix or beta sheet secondary structure and / or are distinct from the helical framework loops and the beta sheets of the antibody scaffold that they join. The linkers, for examples, are of 1-8 residues, 1-7 residues, 1-6 residues, 1-5 residues, 1-4 residues, or 1-3 residues. In some embodiments of any one of the compositions or methods provided herein, the foregoing linkers are at the N-terminus and / or C-terminus of a helical framework loop, such as a helical FR1 or FR3 loop. As another example, the linkers are of 2-8 residues, 2-7 residues, 2-6 residues, 2-5 residues, or 2-4 residues. In some embodiments of any one of the compositions or methods provided herein, the foregoing linkers are at the N-terminus and / or C-terminus of a helical framework loop, such as a helical FR1 loop. In some embodiments of any one of the compositions or methods provided herein, the foregoing linkers are at the C-terminus of a helical framework loop, such as a helical FR3 loop. As yet a further example, a linker can have a length of 1, 2, 3, 4, 5, 6, 7, or 8 residues. Other lengths may be determined by the number of residues that is both sufficient to design binding to a target and that adopts an alpha helical secondary structure with high predicted accuracy and confidence. Still other lengths of a linker, in other embodiments, can be those that result in antibody structures that are less likely to be stable but may still function as desired. Amino acids for the linkers can be, but are not limited to, glycine, proline, and / or serine. Compositions comprising the helical framework loops alone or in combination with the antibody scaffolds as provided herein are provided. Thus, compositions comprising the antibody structures provided herein are provided. For example, the antibody structures comprise any portion of an antibody, full-length or an antigen-binding fragment thereof, with one or more helical framework loops as provided herein. In some embodiments, the antibody structure comprises a VHH scaffold with a helical framework loop, such as a helical FR1 and / FR2 loop; a VHH scaffold with a helical framework loop, such as a helical FR1 and / FR3 loop; a mAb, Fab, or scFv scaffold with a VH helical framework loop, such as a helical FR1 and / FR3 loop, and a VL helical framework loop, such as a helical FR1 and / FR3 loop. a mAb, Fab, or scFv scaffold with a VH helical framework loop, such as a helical FR1 and / FR2 loop, and a VL helical framework loop, such as a helical FR1 and / FR2 loop; a mAb, Fab, or scFv scaffold with a VH helical framework loop, such as a helical FR1 and / FR2 loop, and a VL helical framework loop, such as a helical FR2 loop; a mAb, Fab, or scFv scaffold with a VH helical framework loop, such as a helical FR2 loop, and a VL helical framework loop, such as a helical FR1 and / FR2 loop. Specific examples of antibody structures with folds created by combining helical framework loops and scaffolds as provided herein are depicted in the Figures. For example, FIG.1 depicts (a) a standard nanobody fold and (b) a nanobody fold with helical framework loops of FR1 and loop 3 of FR3. FIG.2 depicts an example nanobody (VHH) fold with helical framework loops FR1 and loop 3 of FR3. FIG.3 depicts an example nanobody (VHH) fold with helical framework loops of FR1 and loop 3 of FR3 (alternate view). The antibody structures provided herein can be designed to have specific features. Successful binding of exemplary antibody structures is demonstrated in the Examples. Further, antibody structures have been designed that avoid undesirable features, such as lack of stability or degradation in a gastric or an intestinal environment and / or lack of stability or degradation by one or more proteases and / or lack of stability or degradation at higher temperatures, such as at 37ºC or above. “Stability” as used herein refers to at least some level of antibody structures in a sample that are still intact and / or have reduced or no unfolding under a set of conditions. Exemplary antibody structures of the Examples are demonstrated to have been successfully produced with stability and degradation resistant features. For example, the antibody structures can be pH resistant or stable at a particular pH. Generally, a pH resistant or stable antibody structure can avoid or have reduced unfolding at a specific pH, such as in acidic conditions, such as the acidic conditions of the stomach. Unfolding can result in reduced efficacy of an antibody. Thus, in some embodiments, the antibody structures provided herein can be resistant to unfolding in such pH, such as acidic conditions. In one embodiment of any one of the compositions or methods provided herein, the antibody structure can be stable or resistant to degradation at a gastric pH, such as a pH of 2. In one embodiment of any one of the compositions or methods provided herein, the antibody structure can be stable or resistant to degradation in simulated gastric fluid, such as the simulated gastric fluid of Table 2. In one embodiment of any one of the compositions or methods provided herein, the antibody structure can be stable or resistant to degradation in simulated gastric fluid for at least 1 hour, 2 hours, 3 hours or 4 hours. In one embodiment of any one of the compositions or methods provided herein, the antibody structure can be stable or resistant to degradation at an intestinal pH, such as a pH of 6.5. In one embodiment of any one of the compositions or methods provided herein, the antibody structure can be stable or resistant to degradation in simulated intestinal fluid, such as the simulated intestinal fluid of Table 3. In one embodiment of any one of the compositions or methods provided herein, the antibody structure can be stable or resistant to degradation in simulated intestinal fluid for at least 1 hour, 2 hours, 3 hours or 4 hours. The antibody structures can be thermostable. Generally, an antibody structure can avoid or have reduced unfolding at higher temperatures and / or have the ability to remain properly folded and maintain its structure when exposed to higher temperatures. The antibody structures with frameworks comprising rigid structures can generally have thermostability relative to (increased stability relative to) conventional antibodies as a property intrinsic to the structures. As an example, the antibody structures provided herein can be thermostable or have resistance to degradation at higher temperatures, such as 37ºC or any one of the temperatures provided herein. De novo designed proteins have been found to be exceptionally thermostable relative to natural proteins. This is due to their rigid, idealized structures and sequences that are a result of how they are designed. As another example, the antibody structures provided herein can be resistant to degradation by proteolytic enzymes, including luminal enzymes from gastrointestinal and pancreatic secretions, bacterial enzymes in the colon and mucosal enzymes. Table 1 provided below lists digestive enzymes, and the antibody structures provided herein can be resistant to degradation by one or more or all (or any combination) of these enzymes. For example, pepsin in the stomach is able to degrade proteins into smaller fragments of peptides by hydrolyzing the peptide bonds. As another example, proteolytic enzymes in the upper part of the small intestine which are secreted by pancreas, such as trypsin, chymotrypsin, carboxypeptidase and elastase, can result in degradation or further degradation of proteins. Moreover, remaining parts of proteins can be finally digested by various peptidases (e.g., aminopeptidase and dipeptidase). Resistance to such degradation can be a benefit. The antibody structures herein can have such resistance in some embodiments. Table 1. Main digestive enzymes and their sites of action Secretion Enzyme Specificity site Stomach Pepsin Asp, hydrophobic amino acids Pancreas Trypsin Arg, Lys Chymotrypsin Aliphatic amino acids (Phe, Tyr) Carboxypeptidase A Aromatic amino acids in C-terminal (Tyr, Phe, Ile, Thr, Glu, His, Ala) Carboxypeptidase B Arg, Lys in C-terminal Elastase Ala, Gly, Ser Small Aminopeptidase A Asp, Glu in N-terminal intestine Aminopeptidase N Ala, Leu in N-terminal Aminopeptidase P Pro in N-terminal Aminopeptidase W Typ, Tyr, Phe in N-terminal Secretion Enzyme Specificity site γ-Glutamyl γ-Glutamic acid in N-terminal transpeptidase Dipeptidyl peptidase IV Pro, Ala Peptidylpeptidase A His–Leu Carboxypeptidase M Lys, Arg in C-terminal Carboxypeptidase P Pro, Gly, Ala in C-terminal γ-Glutamyl γ-Glutamic acid carboxypeptidase Endopeptidase-24.11 Hydrophobic amino acids Endopeptidase-24.18 Aromatic amino acids Enteropeptidase (Asp)4-Lys Stability or resistance to degradation of antibody structures as provided herein can generally be measured by a variety of methods known in the art. For example, Differential Scanning Calorimetry (DSC), Nano-Differential Scanning Calorimetry (nano-DSC), Western Blot, Immunohistochemistry (IHC) and Immunofluorescence (IF), Enzyme-linked immunosorbent assay (ELISA), Dynamic Light Scattering (DLS), Differential Scanning Fluorimetry (DSF), Thermofluor assay, Size-exclusion chromatography (SEC), or Capillary electrophoresis (CE). In one embodiment of any one of the compositions or methods provided herein, the stability or resistance to degradation of an antibody structure is performed by any one of the methods provided herein, such as in the Examples. Stability or resistance to degradation can also be assessed with SDS-PAGE, such as described below. Thus, in one embodiment, the stability or resistance to degradation of an antibody structure of any one of the compositions or methods provided herein can be assessed as follows. 1. Aliquot 7.5 uL of His-tag purified protein sample in elution buffer (25 mM NaPhosphate pH 7.4, 275 mM NaCl, 250 mM Imidazole). 2. Add 7.5 uL of 2x freshly prepared solution (e.g., at a desired pH (depending on whether or not pH degradation is to be assessed)) with or without added proteases at 60 μg / mL, if protease degradation is to be assessed. 3. Incubate at a desired temperature (if degradation at a desired temperature, such as at 37ºC) for 1 hour and 4 hour timepoints. For 0 hour timepoint, add elution buffer to match the volume of the digestion timepoints. 4. Add 5.5uL of 4x Tris / Glycine / SDS running buffer (BioRad) and 1uL 1M DTT to each sample and heat 5 minutes at 95C to end incubation. 5. Run samples immediately on 4-20% TGX gel (BioRad) to perform SDSPAGE analysis using Precision Plus Protein Dual Xtra Prestained Protein Standards (BioRad) as ladder. Thermostability can be measured by the melting temperature, at which point the protein denatures. Thermostability can correlate with resistance to proteolysis as well as other metrics of stability, such as resistance to guanidinium denaturation. Thermostability can also be measured by any other method as provided herein or as otherwise known in the art. Thermostability is a highly desirable characteristic, in some embodiments, as it can enable manufacturing, storage, and longer serum half-life. In one embodiment of any one of the compositions or methods provided herein, the antibody structures provided herein can be thermostable and have any one of the foregoing thermostable features. As another example, the polypeptides provided herein can be orally delivered and / or shelf stable. “Orally deliverable” and the like refers to an antibody structure, a polypeptide or composition comprising the antibody structure or polypeptide that can be delivered by oral means to a subject and results in at least some in vivo activity and / or efficacy. “Shelf stable” refers to the ability to store the antibody structure, polypeptide or composition comprising the antibody structure or polypeptide at at least room temperature or any one of the temperatures provided herein without significant degradation and / or reduction in efficacy. Still other properties include, but are not limited to, features such as pH responsiveness. In one embodiment, an antibody structure may bind with high affinity at a neutral or slightly basic pH (such as physiologic pH of 7.4) while binding with much lower affinity in an acidic environment (such as pH of 4.5 to 6.5 such as of the endosome). In such an embodiment, pH responsiveness can be used to increase internalization of an antibody structure by promoting its dissociation in the endosome. Other environments can also feature an acidic pH, like the tumor microenvironment (TME). A pH responsive antibody structure may bind with high affinity in the acidic TME, while featuring a much lower affinity in neutral physiologic pH. In such an embodiment, pH responsiveness can be used to limit an antibody structure to only bind its target in the TME, increasing its specificity. In one embodiment of any one of the compositions or methods provided herein, the antibody structures provided herein can be pH responsive. In one embodiment of any one of the compositions or methods provided herein, the antibody structures provided herein can be pH responsive and have any one of the foregoing pH responsive features. Histidine residues have a pKa of around 6, so they become protonated in acidic environments and may alter binding affinity if they are present at a binding interface. In one embodiment of any one of the compositions or methods provided herein, a pH responsive antibody structure can feature histidine residues at the binding interface. These histidine residues may be on the paratope of the antibody structure and / or on the epitope of the target of the antibody structure. The compositions provided herein can further comprise a pharmaceutically acceptable carrier. As used herein, “pharmaceutically acceptable carrier” includes any and all salts, buffering agents, preservatives, compatible carriers, solvents, dispersion media, coatings, antibacterial and antifungal agents, isotonic and absorption delaying agents, and the like that are physiologically compatible. A pharmaceutically acceptable carrier also includes one or more compatible solid or liquid fillers, diluents or encapsulating substances that are suitable for administration into a human. The term “carrier” denotes an organic or inorganic ingredient, natural or synthetic, with which the active ingredient is combined to facilitate the application. A carrier may be suitable for oral administration. In some embodiments, a composition may conveniently be presented in unit dosage form and may be prepared by any of the methods well-known in the art of pharmacy. In some embodiments, compositions are prepared by uniformly and intimately bringing the active compound into association with a liquid carrier, a finely divided solid carrier, or both, and then, if necessary, shaping the product. Compositions suitable for administration may comprise a sterile aqueous or non-aqueous preparation. This preparation may be formulated according to known methods using suitable dispersing or wetting agents and suspending agents. The sterile injectable preparation also may be a sterile injectable solution or suspension in a non-toxic parenterallyacceptable diluent or solvent. Among the acceptable vehicles and solvents that may be employed are water, Ringer's solution, and isotonic sodium chloride solution. In addition, sterile, fixed oils are conventionally employed as a solvent or suspending medium. For this purpose any bland fixed oil may be employed including synthetic mono- or di-glycerides. In addition, fatty acids such as oleic acid may be used in the preparation of injectables.Carrier formulations suitable for administration can be found in Remington's Pharmaceutical Sciences, Mack Publishing Co., Easton, Pa. Any of the compositions provided herein may be sterile. Compositions as provided herein, in some embodiments, may be administered in effective amounts to a subject. An “effective amount” is that amount of an active compound that alone, or together with further doses, produces a desired response. Such effective amounts will depend, of course, on the particular condition being treated, the severity of the condition, the individual patient parameters including age, physical condition, size and weight, the duration of the treatment, the nature of concurrent therapy (if any), the specific route of administration and like factors within the knowledge and expertise of the health practitioner. These factors are well known to those of ordinary skill in the art and can be addressed with no more than routine experimentation. It is generally preferred that a maximum dose of the individual components or combinations thereof be used, that is, the highest safe dose according to sound medical judgment. It will be understood by those of ordinary skill in the art, however, that a patient / subject may insist upon a lower dose or tolerable dose for medical reasons, psychological reasons or for virtually any other reason. The doses of compositions administered to a subject can be chosen in accordance with different parameters, in particular in accordance with the mode of administration used and the state of the subject. Other factors include the desired period of treatment. Compositions as provided herein, in some embodiments, may be administered in effective amounts to a subject. "Administering" or "administration" or “administer” means providing a material to a subject in a manner that is pharmacologically useful. The term is intended to include “causing to be administered”. “Causing to be administered” means causing, urging, encouraging, aiding, inducing or directing, directly or indirectly, another party to administer or ingest the material. An “effective amount” is that amount of an active compound that alone, or together with further doses, produces a desired response. Such effective amounts will depend, of course, on the particular condition being treated, the severity of the condition, the individual patient parameters including age, physical condition, size and weight, the duration of the treatment, the nature of concurrent therapy (if any), the specific route of administration and like factors within the knowledge and expertise of the health practitioner. These factors are well known to those of ordinary skill in the art and can be addressed with no more than routine experimentation. It is generally preferred that a maximum dose of the individual components or combinations thereof be used, that is, the highest safe dose according to sound medical judgment. It will be understood by those of ordinary skill in the art, however, that a patient / subject may insist upon a lower dose or tolerable dose for medical reasons, psychological reasons or for virtually any other reason. The doses of compositions administered to a subject can be chosen in accordance with different parameters, in particular in accordance with the mode of administration used and the state of the subject. Other factors include the desired period of treatment. As used herein, the term “subject” is intended to include humans and non-human animals, including warm blooded mammals, such as humans and primates; avians; domestic household or farm animals such as cats, dogs, sheep, goats, cattle, horses and pigs; laboratory animals such as mice, rats and guinea pigs; fish; reptiles; zoo and wild animals; and the like. In one aspect, a method of administering any one of the antibody structures or any one of the compositions provided herein to a subject is provided. In one embodiment, the administering is done by oral delivery. Methods of Producing Antibody Structures with Helical Framework Loop(s) The methods provided herein allow the antibody structures comprising at least one helical framework loop on a desired antibody scaffold to be designed by, for example, inputting structural requirements and / or a target sequence. The target input structure can, for example, a crystal structure or a predicted structure. The resulting antibody can take a variety of formats, including nanobody and scFv. Antibodies with helical framework loop(s) can be designed by first generating a protein backbone and then generating an amino acid sequence for that backbone. The resulting designs can be, optionally, evaluated with computational metrics. The antibody backbones can be generated in a variety of ways. For example, backbones for antibodies with at least one helical framework loop can be generated computationally using a diffusion model, such as a denoising diffusion probabilistic model. Such a model, generally, is a generative model that outputs backbone atom coordinates for a protein. Such models can be used with fold conditioning, which in some embodiments allows for the specification of a protein fold. In some embodiments, the fold can be specified by providing secondary structural elements, such as each secondary structural element that is desired of the desired antibody structure (e.g., alpha helix, beta sheet, or loop), total length, or length of specific elements, in number of residues, and / or contacts with other secondary structural elements. As another example, backbones can be designed to bind to a target by providing a model with a target structure and desired residues on that structure (e.g., the epitope). Backbones for antibodies with helical framework loop(s) can be generated in reference to binding to a target by, for example, a grafting approach. In one example of such an approach, first, a backbone for an antibody with helical framework loop(s) is generated without providing a target, using a diffusion model with fold conditioning to specify the fold. A backbone with one, two, three or four helices in a helical bundle fold can then be generated that binds a target using a diffusion model with fold conditioning as well as a target structure and target residues on that structure. The helices can then grafted onto the antibody with helical framework loop(s) backbone. In one example, a method can be performed comprising, A) 1) selecting a target structure and, optionally, epitope residues (e.g., an epitope) on the structure, 2) using a diffusion model (e.g., RoseTTAFold diffusion model) with fold conditioning to generate an antibody structure to the target and, optionally, the epitope residues (e.g., epitope), 3) providing a backbone comprising one, two, three or four helices to specify a fold, and 4) using a model (e.g., ProteinMPNN model) to determine a sequence for the antibody structure, and 5) optionally, using a model (e.g., AlphaFold2) to predict and / or evaluate the binder and / or binding interface; B) 1) using a diffusion model with fold conditioning to generate a backbone with helical framework loop(s), such as by providing a blueprint specifying a fold, 2) using a model (e.g., ProteinMPNN model) to determine a sequence for the backbone with helical framework loop(s); and 3) optionally, using a model (e.g., AlphaFold2) to predict and / or evaluate the structure; and C) 1) grafting the backbone with helical framework loop(s) onto the antibody structure (e.g., by using structural alignment). The foregoing may also comprise redesigning one or more of the residues using a model (e.g., ProteinMPNN) to accommodate the grafting and / or using a model (e.g., AlphaFold2) to predict and / or evaluate the antibody structure with helical framework loop(s) binding to the target. As a further example, a whole-antibody approach can be taken. An example of such approach comprises generating a backbone for an antibody structure with helical framework loop(s) using a target structure and target residues on that structure and using a diffusion model with fold conditioning. In one embodiment, a method can be performed comprising 1) selecting a target structure and, optionally, epitope residues (e.g., an epitope) on the structure; 2) using a diffusion model (e.g., RoseTTAFold diffusion model) with fold conditioning to generate an antibody structure with one, two, three, or four helical framework loop(s) that binds to the target and, optionally, the epitope residues; 3) using a model (e.g., ProteinMPNN model) to determine a sequence for the antibody with helical framework loop(s); and 4) optionally, using a model (e.g., AlphaFold2) to predict and / or evaluate the structure and / or binding interface. The foregoing method may also comprise using a diffusion model with fold conditioning to generate a backbone with helical framework loop(s), such as by providing a blueprint specifying a fold. The result of the foregoing approaches can be a designed backbone for an antibody structure with helical framework loop(s) in complex with a target structure, docked at specified target residues on that structure. The methods provided herein can also include a step of generating amino acid sequences, which step can occur before or after the generation of the backbone. Amino acid sequences can be generated, for example, using a message-passing neural network model. For example, the sequences for a target and an antibody scaffold can be provided to the model along with the backbone structure of the complex (consisting of target and antibody with helical framework loop(s)), and a sequence is generated for the helical framework loop(s) in this context. When a grafting approach is employed, the sequences can be designed either before or after the helices are grafted onto the antibody structure with helical framework loop(s). If necessary for improving predicted accuracy and confidence, some of the amino acids of the antibody scaffold sequence may be redesigned to better support the helical framework loop(s). In particular, the amino acids neighboring the helical framework loop(s) may be redesigned. In some embodiments, optional partial diffusion can be used to refine and generate more designs. The diffusion model may be used in a ‘partial diffusion’ mode to redesign a designed antibody structure with helical framework loop(s). This method can involve adding some noise to the backbone coordinates, then using the diffusion model to generate a new backbone from the ‘noisy’ backbone. A new sequence is then generated for this new backbone. This method can provide additional structural variation and yield improved designs. Designs generated as provided herein can be evaluated in some embodiments. Protein structure prediction models may be used. Such models can provide a predicted structure given the sequence of the design, which can be aligned to the design structure model to assess accuracy. These models also can provide a measure of confidence in the prediction. The methods provided herein can include steps to design antibody structures with the specific feature(s) provided herein. Any one of the methods provided herein, can include one or more steps to design an antibody structure to have one or more features as provided herein. As an example, a pH responsive antibody structure can be designed to target an epitope featuring histidine residues. One or multiple histidine residues can be provided on the target as target residues. The histidine residues can be modeled in either their protonated or deprotonated state. Generated antibody structures can be selected that make interactions with either the protonated or deprotonated histidine residue(s), depending on the desired pH dependent binding behavior. As another example, a pH responsive antibody structure can be designed to incorporate histidine residues when generating the sequences of the framework loops comprising a rigid structure. These histidine residues can be preferentially used in the sequence design, such as by providing amino acid composition bias weights. These weights bias the sequence design process to use histidine residues with a higher probability when designing the interface. The histidine residues can be modeled in either their protonated or deprotonated state. Generated antibody structures comprising a rigid structure can be selected that feature one or more histidine residues in their loops comprising a rigid structure. These antibody structures may make interactions with the target using either the protonated or deprotonated histidine residue(s), depending on the desired pH-dependent binding behavior. Any one of the methods provided herein, can include any one or more of the foregoing steps to design an antibody structure to be pH responsive. Any one of the methods provided herein, can include one or more steps to select or test for pH responsiveness. Any one of the methods provided herein, can include one or more steps to design an antibody structure to be stable or resistant to degradation at a specific pH, such as a pH of 2 or 6.5. Any one of the methods provided herein, can include one or more steps to select or test for pH resistance, such as at a pH of 2 or 6.5. Any one of the methods provided herein, can include one or more steps to design an antibody structure to be thermostable or be resistant to degradation at higher temperatures, such as 37ºC or any one of the temperatures provided herein. Any one of the methods provided herein, can include one or more steps to select or test for thermostability or resistance to degradation at higher temperatures, such as 37ºC or any one of the temperatures provided herein. The methods provided herein can include steps to design antibody structures that have thermostability and / or be shelf stable. Any one of the methods provided herein, can include one or more steps to design an antibody structure to be thermostabile and / or shelf stable. Any one of the methods provided herein, can include one or more steps to select or test for thermostability and / or be shelf stability. Any one of the methods provided herein, can include one or more steps to design an antibody structure to be stable or resistant to degradation by one or more proteases. Any one of the methods provided herein, can include one or more steps to select or test for proteolytic stability or resistance or resistance to degradation by one or more proteases. Any one of the methods provided herein, can include one or more steps to design an antibody structure to be stable or resistant to degradation in a gastric environment. Any one of the methods provided herein, can include one or more steps to design an antibody structure to be stable or resistant to degradation in simulated gastric fluid, such as that of Table 2. Any one of the methods provided herein, can include one or more steps to select or test for stability or resistance to degradation in a gastric environment. Any one of the methods provided herein, can include one or more steps to select or test for stability or resistance to degradation in a simulated gastric fluid, such as that of Table 2. Any one of the methods provided herein, can include one or more steps to design an antibody structure to be stable or resistant to degradation in an intestinal environment. Any one of the methods provided herein, can include one or more steps to design an antibody structure to be stable or resistant to degradation in simulated intestinal fluid, such as that of Table 3. Any one of the methods provided herein, can include one or more steps to select or test for stability or resistance to degradation in an intestinal environment. Any one of the methods provided herein, can include one or more steps to select or test for stability or resistance to degradation in a simulated intestinal fluid, such as that of Table 3. The methods provided herein can include steps to design antibody structures to be delivered orally. Any one of the methods provided herein, can include one or more steps to design an antibody structure to be orally deliverable. Any one of the methods provided herein, can include one or more steps to select or test for oral deliverability. The present invention is further illustrated by the following Examples, which in no way should be construed as further limiting. The entire contents of all of the references (including literature references, issued patents, published patent applications, and co pending patent applications) cited throughout this application are hereby expressly incorporated by reference. However, the citation of any reference is not intended to be admission that said reference is prior art. EXAMPLES Example 1 – Example nanobody with two helical framework loops designed to bind to IL7Rα (generated using docking approach) A nanobody backbone with helical framework loops was generated with IL7Rα as the target structure and with target residues on IL7Rα inputted into a diffusion model with fold conditioning. The IL7Rα target sequence is as follows: DYSFSCYSQLEVNGSQHSLTCAFEDPDVNTTNLEFEICGALVEVKCLNFRKLQEIYFIETKKF LLIGKSNICVKVGEKSLTCKKIDLTTIVKPEAPFDLSVVYREGANDFVVTFNTSHLQKKYVK VLMHDVAYRQEKDENKWTHVNLSSTKLTLLQRKLQPAAMYEIKVRSIPDHYFKGFWSEWS PSYYFRTP The designed nanobody with two helical framework loops sequence (helical framework loops in bold) is as follows: ATITVEAADKTIEYESPQDLVAEVINKGSLTINLTVTITTDTSANIALAVEVTIGDKTYKALFL ISSDGSVTPLESSPIVSVSATKNGNTITVTITINIPKEDAFEMILKGNTTITVKVAANLYEEGM TAEDVFSEATSSVTATITFKFK An evaluation of the designed nanobody was performed. The computational metrics were as follows: i. AlphaFold2 predicted local distance difference test (pLDDT): 91.1 ii. AlphaFold2 interaction predicted aligned error (pAE): 9.478 iii. AlphaFold2 monomer root mean-squared deviation (RMSD): 0.892Å With these metrics, this design can be considered an in silico success, for example, according to References 5 and 7: AlphaFold2 monomer pLDDT > 80, AlphaFold2 interaction pAE < 10, AphaFold2 monomer RMSD < 1Å. FIG.4 below shows the resulting designed nanobody; design model of complex aligned to predicted structure of complex. FIG.5 shows the designed nanobody with interface detail of the predicted structure. Example 2 – Example nanobodies with helical framework loop(s) designed to bind to IL7Rα Nanobody backbones with helical framework loops were generated with IL7Rα as the target and with methods provided herein. Sequences for exemplary polypeptides are provided below. Full name Sequence w2tn3_qgxd6_4_1_5_13_6_5_20_5_21_11_22_4_w6 ETQTLTSTDVLLSEEEVAEMIIKAIKGEPFSIRVSCAV lk4 SGEFLPSDEISCGVQLPGKGLEWVSAYNVGTGETY YADSSGLRATVSVDNSKNTVYCQLTFKTKEKVAEFI SLLYSGESVVVTCALLRDGKLVGQGTAQIKLS 3dkxo_3m7vt_4_0_5_10_6_5_20_5_21_9_22_5_hyt3 EVQVLVSDVTLSAEELAKETYKTGKLIIRVSCAVSSS q SPEPLTVSVVLQAPGGGLLGSGAGNLATGETYNAD SSKWRVTLSFDNSKGTVTAQFETKGDVADSILLLRE LKGPLVVYCAAYVNGELAGQASGKTTIS 71t3a_5clcq_4_0_5_9_6_5_20_5_21_10_22_5_7xwp ETQVLVSSPVLTKEEVLKQIESGELTIRSSCAVSSSE m DKPITVSCVLQAGGGGLLASAAVSSSQSSTSGAGS DGGRLTVSYDNSKGTVTCQFSYKGPKVEVAELVQ DIIYNGAVVYCAAYEDGKLVAQDSAPIVFS 71t3a_5clcq_4_0_5_9_6_5_20_5_21_10_22_5_9fehr ETQVLVSSPVLTEKEVIEQIEKGVFKIRSSCAVSSSE DKPITVSCVLQAGGGGLLASAAVSSSQSSTSGAGS DGGRLTVSYDNSKGTVTCQFTIEDDPVKVAEIVQE MIFNGAVVYCAAYEDGKLVAQDSAPIVLS 71t3a_5clcq_4_0_5_9_6_5_20_5_21_10_22_5_spg1 ETQVLVSSPVLTEEEIKEGIEKGEFVIRSSCAVSSSE 1 DKPITVSCVLQAGGGGLLASAAVSSSQSSTSGAGS DGGRLTVSYDNSKGTVTCQFTLKGSKAKVAEFVED LYFNGAVVYCAAYEDGKLVAQDSAPIVFS 71t3a_5clcq_4_0_5_9_6_5_20_5_21_10_22_5_tmzu ETQVLVSSPVLTEEEVIESAKKGELTIRSSCAVSSSE z DKPITVSCVLQAGGGGLLASAAVSSSQSSTSGAGS DGGRLTVSYDNSKGTVTCQFVKKAPPVEMAELLM EMIFNGAVVYCAAYEDGKLVAQDSAPIVFS 71t3a_5clcq_4_0_5_9_6_5_20_5_21_10_22_5_z7ey ETQVLVSSPVLTEEEVIKQIESGVLTIRSSCAVSSSE 1 DKPITVSCVLQLGGGGLLASAAVSSSQSSTSGAGS DGGRLTVSYDNSKGTVTCQFTYKGSPVEVAELVQE IKYEGAVVYCAAYEDGKLVAQDSAPIVFS 78m01_qgxd6_4_1_5_13_6_5_20_5_21_11_22_4_2l ETQTLTSTDVLLSKEEFLEMIDKTLLGEPISIRVSCAV sek SGELLPSDEISCGVQLPGKGLEWVSAYNVGTGETY YADSSGLRATVSVDNSKNTVYCQLTFTTRENQEAV LLSLYKGEPVVVTCALLRDGKLVGQGTAQIKLS 78m01_qgxd6_4_1_5_13_6_5_20_5_21_11_22_4_g ETQTLTSTDVLLTDEELKKMIDEFHEGKPLKIRVSCA 5d9p VSGEFLPSDEISCGVQLPGKGLEWVSAYNVGTGET YYADSSGLRATVSVDNSKNTVYCQLTATTRETSEEI LLLLAEGKPVVVTCALLRDGKLVGQGTAQIKLS 9cjwm_qgxd6_4_1_5_13_6_5_20_5_21_11_22_4_qg ETQTLTSTDVLLNEEEFEELVLKAILGEPISIRVSCAV 166 SGELLPSDEISCGVQLPGKGLEWVSAYNVGTGETY YADSSGLRATVSVDNSKNTVYCQLTTKTKEESEAVI TALASGKPVVVTCALLRDGKLVGQGTAQIKLS a7a1n_lqjum_4_0_5_9_6_5_20_4_21_13_22_4_fl6u8 EVQVLASDVTLSLDDVVAALHRGEPLFRVSCAVSA DDVASLTVTCQLQAPGGGLLATATASASQSGTSAA GSAGFRVTVSVDVSKGTATCQVSAADREKAEKAIIE AVEAGGKFTVVCTATDGGETATATAQVTLSS a7a1n_pdyub_4_0_5_9_6_5_20_5_21_12_22_4_afxv EVQVLASDVTLSAQEIVDKVLSGEKTLRVSCAVSGE 4 LLPDTVVSCQVQAPGGGLLGSAAVSASQSSGYGA DSPKGRVTVSVDNSKNTVYCQLEFKDVYTAVDLIH ECLQSGKGLVVYCAVSSGGETWGAGGQITISS a7a1n_pdyub_4_0_5_9_6_5_20_5_21_12_22_4_q6h EVQVLASDVTLSAEDVAKKINEGNTTLRVSCAVSGE 94 LLPDTVVSCQVQAPGGGLLGSGAVSASQSSGYGA DSPKGRVTVSVDNSKNTVYCQLHFKDKETFIDLVH EMLITGKGLVVYCAVSSGGETWGAGGQITISS a7a1n_pdyub_4_0_5_9_6_5_20_5_21_12_22_4_qhyi EVQVLASDVTLSSDEIVEKIQAGDTTLRVSCAVSGE x LLPDTVVSCQVQAPGGGLLGSGAVSASQSSGYGA DSPKGRVTVSVDNSKNTVYCQLEFANREQFIEMVR ECLISGKGLVVYCAVSSGGETWGAGGQITISS a7a1n_qgxd6_4_1_5_13_6_5_20_5_21_11_22_4_01 ETQTLTCTDVLLSEEEFEEMRLKAIKGEPITIRVSCA rzs VSGEFLPSDEISCGVQLPGKGLEWVSAYNVGTGET YYADSGGLRATVSVDNSKNTVYCQLTVEGKEEQE EILKLLAKGEGVVVTCALLRDGKLVGQGTAQIKLS a7a1n_qgxd6_4_1_5_13_6_5_20_5_21_11_22_4_3g ETQTLTSTDVLLSEEELEELRIKAIHGEPISIRVSCAV g9b SGEFLPSDEISCGVQLPGKGLEWVSAYNVGTGETY YADSSGLRATVSVDNSKNTVYCQLTFEKKEDAEKF VKAILSGKPVVVTCALLRDGKLVGQGTAQIKLS a7a1n_qgxd6_4_1_5_13_6_5_20_5_21_11_22_4_h0 ETQTLTSTDVLLTEEEFEEMRIKAIHGEPVSIRVSCA 5qm VSGEFLPSDEISCGVQLPGKGLEWVSAYNVGTGET YYADSSGLRATVSVDNSKNTVYCQLEFLTKEEAEE AIKLLAKGEPVVVTCALLRDGKLVGQGTAQIKLS a7a1n_qgxd6_4_1_5_13_6_5_20_5_21_11_22_4_hk ETQTLTSTDVLLNEEQFEEYRLKAIRGEPITIRVSCA p1t VSGEFLPGDEISCGVQLPGKGLEWVSAYNVGTGET YYADSSGLRATVSVDNSKNTVYCQLSAKTPEASKEI IKRLANGEPVVVTCALLRDGKLVGQGTAQIKLS n7wni_qgxd6_4_1_5_13_6_5_20_5_21_11_22_4_kq ETQTLTSTDVLLSEKELKEMIDKAFAGEPIEIRVSCA mo3 VSGEFLPSDEISCGVQLPGKGLEWVSAYNVGTGET YYADSSGLRATVSVDNSKNTVYCQLTVKDPKLALE MLELMYNGEPVVVTCALLRDGKLVGQGTAQIKLS n7wni_qgxd6_4_1_5_13_6_5_20_5_21_11_22_4_za ETQTLTSTDVLLNEKELKEFVDKAFSGEPVTIRVSC 15i AVSGEFLPSDEISCGVQNPGKGLEWVSAYNVGTG ETYYADSSGLRATVSVDNSKNTVYCQLIFETTEDAL QFLQDVYDGTPVVVTCALLRDGKLVGQGTAQIKLS oiwi5_794f9_4_0_5_9_6_5_20_4_21_11_22_4_bg52 EVQVLVSDTTLSLDELREELLKSGGKLRYSCAVSTS 6 TPAPVTVSWYLQAPGGGLTLVGTASAPGATGTQA VSPKLRGTVSNDGSSGTATLTLEFKDEDAAAEFVLL LLKEPLVLTCAATQGGSTATATGQLTVSS oiwi5_l52uq_4_0_5_9_6_4_20_4_21_12_22_5_er2oj EIQVLVSPVTLTTEEFIEKVVLGEPIRVSCAISGTLQP GTRVTCVWQAPGGGLLGQAAVTVGGGSTVLADSP KYRVTVSVDNSKNTAYCQLTAVKREDWLEVVNTMI NEGKGLVVTCTVEDGGDTDTQTGQVTLSS oiwi5_qgxd6_4_1_5_13_6_5_20_5_21_11_22_4_j2ds ETQTLTSTDVLLTDEEIEELRLKVIKGEPFSIRVSCAV 6 SGELLPSDEISCGVQLPGKGLEWVSAYNVGTGETY YADSSGLRATVSVDNSKNTVYCQLTATKKEAGAEF VKLLVNGEPVVVTCALLRDGKLVGQGTAQIKLS oiwi5_qgxd6_4_1_5_13_6_5_20_5_21_11_22_4_ofr1 ETQTLTSTDVLLSEEEFEEMLSKLLNGEPVSIRVSC 5 AVSGEFLPSDEISCGVQLPGKGLEWVSAYNVGTGE TYYADSSGLRATVSVDNSKNTVYCQLSGKTREAW EEFLKTLAEGKSVVVTCALLRDGKLVGQGTAQIKLS rwka4_5clcq_4_0_5_9_6_5_20_5_21_10_22_5_yug2 ETQVLVSSPVLTREEMIEQIKSGVFTIRSSCAVSSSE 4 DKPITVSCVLQAGGGGLLASAAVSSSQSSTSGAGS DGGRLTVSYDNSKGTVTCQFIYEGSSEEIQYLLTRL YRNPAVVYCAAYEDGKLVAQDSAPIVFS rwka4_qgxd6_4_1_5_13_6_5_20_5_21_11_22_4_ba ETQTLTSTDVLLTEEELKELAKKAALGEPAVIRVSCA 25d VSGEFLPGDEISCGVQLPGKGLEWVSAYNVGTGET YYADSSGLRATVSVDNSKNTVYCQLSAKTPELMQE ILLRVYAGEPVVVTCALLRDGKLVGQGTAQIKLS s47jh_qgxd6_4_1_5_13_6_5_20_5_21_11_22_4_ibqi ETQTLTSTDVLLTDEEFEEMKLKVFEGKPISIRVSCA q VSGELLPSDEISCGVQLPGKGLEWVSAYNVGTGET YYADSSGLRATVSVDNSKNTVYCQLTLNTKEAAEE FLKCLADGKGVVVTCALLRDGKLVGQGTAQIKLS w2tn3_qgxd6_4_1_5_13_6_5_20_5_21_11_22_4_6b ETQTLTSTDVLLSEDDLVEMRLNVLKGEPITIRVSCA gu7 VSGEFLPSDEISCGVQLPGKGLEWVSAYNVGTGET YYADSSGLRATVSVDNSKGTVYCQLTVTDKETAIEII KNLMAGKPVVVTCALLRDGKLVGQGTAQIKLS Example 3 – Example nanobodies with helical framework loop(s) designed to bind to IL23α Nanobody backbones with helical framework loops were generated with IL23α as the target and with methods provided herein. Sequences for exemplary polypeptides are provided below. Full name Sequence 52gmr_093gi_4_0_5_9_6_4_20_5_21_11_22_4_1dll ELQVLVSDVTLSQEELYELFRSGKPIRVSCAISGPVSPS p ATVSCVWQAGGGGLLASAAAKFSESSTYAAGSSGVRL TVSVDNSKNTVYCQLQVEGGPWLADKVFALAVLEGLV VTCTASDGGDTATAGGQVTISS 52gmr_fv8m7_4_0_5_9_6_4_20_4_21_12_22_4_ja3 ERQVLVSDVTLTRAEAGERILAGEGLRSSCAVSGELQP y9 SDVVSCSWQAAGGGLLGSAAASASESSTTAAGSAGF RVTVSIDNSKGTVTCSLSPSDLEARIRLLALIERDPVLVV TCAVLRNGELVGQASGRITLS 52gmr_gmta4_4_0_5_10_6_4_20_4_21_9_22_4_qu ETQVLASDTTLSADEVFANAYAGKPATRTSCATSSTLH nmz PDDVVSCVAQAPGKGLEASGAANVGTGETYYGDSAK GRVTVSVDNSKNTFYCQFYFTHLLDCALLLASGDDVVF YCAVIRGGKVWGQGSATTKFS 52gmr_iusee_4_0_5_9_6_4_20_5_21_12_22_5_ety EIQVTVSDVTLSAEEVMEKALRGEPIRVSCAVSGTLTSS 2t TVATVVWQAPGGGLLGQSAATSASGSTTLADSPVFRV TASIDNSKNTVTGTLSARASPWAFAVLLAAIIERSGVLT VTCTITDGSDTSTSTGQLTISS 52gmr_jdnu4_4_1_5_9_6_4_20_4_21_11_22_5_3jy EAQVLSVSGVLLPISEFGNLLLRGEGIRVSCAVSGPLQP b4 STLVSCQLQAAGGGLIGSAAAKAGQTGTYYSDSTGVR VTVSVDNSKNTAYCQVKASNREALVKLLYLMQKYKSFV VYCAVTDGGKTSGQGGQITISS 52gmr_l52uq_4_0_5_9_6_4_20_4_21_12_22_5_ajw EIQVLVSPVTLTMDEVMALALAGEGIRVSCAISGTLQPG 0e TRVTCVWQAPGGGLLGQAAVTVGGASTVLADSPKYR VTVSVDNSKNTAYCQLSGREPWYLLALADQALRSGKG LVVTCTVEDGGDTDTQTGQVTLSS 52gmr_l52uq_4_0_5_9_6_4_20_4_21_12_22_5_ge EIQVLVSPVTLTMDEVLALIQAGEPIRVSCAISGTLQPGT 4qo RVTCVWQAPGGGLLGQAAVTVGGASTVLADSPKYRV TVSVDNSKNTAYCQLSTTEPVLMLALAHRALRSGRGLV VTCTVEDGGDTDTATGQVTFSS 52gmr_l52uq_4_0_5_9_6_4_20_4_21_12_22_5_qr7 EIQVLVSPVTLTMDEVLALALAGEGIRVSCAISGTLQPG 7r TRVTCVWQAPGGGLLGQAAVTGGGASTVLADSPKYR VTVSVDNSKNTAYCQLLGERPIYLLALAHQAVASGRGL VVTCTVEDGGDTDTQTGQVTVSS 52gmr_l52uq_4_0_5_9_6_4_20_4_21_12_22_5_uq EIQVLVSPVTLTMEEVLALLAAGQGIRVSCAISGTLQPG obi TRVTCVWQAPGGGLLGQAAVTVGGASTVLADSPKYR VTVSVDNSKNTAYCQLHAAEPWEMLALADEALRSGKG LVVTCTVEDGGDTDTQTGQVTVSS 52gmr_l52uq_4_0_5_9_6_4_20_4_21_12_22_5_v7z EIQVLVSPVTLTMDEVMALALAGEPIRVSCAISGTLQPG fn TRVTCVWQAPGGGLLGQAAVTVGGASTVLADSPKYR VTVSVDNSKNTAYCQLSSREPYALLAEADRALRSGRG LVVTCTVEDGGDTGTQTGQVTLSS 52gmr_l52uq_4_0_5_9_6_4_20_4_21_12_22_5_vl7 EIQVLVSPVTLTMDEVMALALAGEPIRVSCAISGTLQPG 0a TRVTCVWQAPGGGLLGQAAVTVGGASTVLADSPKYR VTVSVDNSKNTAYCQLVAARPEDLLALAYLALRSGKGL VVTCTVEDGGDTDTQTGQVTLSS dzsi2_lqjum_4_0_5_9_6_5_20_4_21_13_22_4_amv EVQVLASDVTLSFTEAWNRLAAGEPIFRVSCAVSADDV 0x ASLTVTCVLQAPGGGLLATATASASQSSTSAAGSAGFR VTVSVDVSKGTATCQVVAADKRRTALALLEAMAAGGK FTVVCTATDGGETATATAQVTLSS dzsi2_lqjum_4_0_5_9_6_5_20_4_21_13_22_4_c1oz EVQVLASDVTLSFTEALNRVAAGEPVFRVSCAVSADDV z ASLTVTCVLQAPGGGLLATATASASQSSTSAAGSAGFR VTVSVDVSKGTATCQVTAADRRAAYLTLLRAYHESGLF TVVCTATDGGETATATAQVTLSS dzsi2_qgxd6_4_1_5_13_6_5_20_5_21_11_22_4_qp ETQTLTSTDVLLSEEQLVALWERTVRGEPATIRVSCAV lt9 SGELLPSDEISCGVQLPGKGLEWVSAYNVGTGETYYA DSSGLRATVSVDNSKNTVYCQLIITDARDMLRLILLLISG QNVVVTCALLRDGKLVGQGTAQIKLS lwgrp_qgxd6_4_1_5_13_6_5_20_5_21_11_22_4_qg ETQTLTSTDVLLTEQEVADILLAAFRGEPATIRVSCAVS bvz GEFLPSDEISCGVQLPGKGLEWVSAYNVGTGETYYAD SSGLRATVSVDNSKNTVYCQLTIEDPELALRFALALIRG EPVVVTCALLRDGKLVGQGTAQIKLS wychj_1f6lv_4_0_5_10_6_4_20_4_21_10_22_4_omr ELQVSVSDVTLSQTDVVNEWLANGGSVRFSCAISGPL d2 SPDAVVSCVLQDAGGGLEASVAFKVSESSASAAGSW GGRLTVSVDASKGTVTCVVSAANLAQFQLFMEALLAGL VLTCTASSNGDTATASGQVTASS wychj_l52uq_4_0_5_9_6_4_20_4_21_12_22_5_tchv EIQVLVSPVTLTPQEVLRLIASGEPIRVSCAISGTLQPGT v RVTCVWQAPGGGLLGQAAVTVGGASTVLADSPKYRV TVSVDNSKGTAYCQLSAARGEALLAELARGYLSGRGL VVTCTVEDGGDTDTQTGQVTLSS Example 4 – Example nanobodies with helical framework loop(s) designed to bind to TNFα Nanobody backbones with helical framework loops were generated with TNFα as the target and with methods provided herein. Sequences for exemplary polypeptides are provided below. Full name Sequence 20cjl_qgxd6_4_1_5_13_6_5_20_5_21_11_22_4_7 ETQTLTSTDVLLNKNQLRAMQRKALLGEPISIRVS 26dn CAVSGEFLPSDEISCGVQLPGKGLEWVSAYNVG TGETYYADSSGLRATVSVDNSKNTVYCQLTATTP EAAQEFVNLLLAGQSVVVTCALLRDGKLVGQGT AQIKLS 20cjl_qgxd6_4_1_5_13_6_5_20_5_21_11_22_4_i1 ETQTLTSTDVLLTRNELTRLQRAAILGEPVTIRVS q51 CAVSGEFLPSDEISCGVQLPGKGLEWVSAYNVG TGETYYADSSGLRATVSVDNSKGTVYCQLTLSTP EALAAFAAALLAGQPVVVTCALLRDGKLVGQGTA QIKLS 44btv_jdnu4_4_1_5_9_6_4_20_4_21_11_22_5_3y EAQVLSLSGVLLTRAEVWERLRAGEPIRVSCAVS 25a GPLQPSTLVSCQLQAAGGGLIGSAAAKAGQTGT YYSDSTGVRVTVSVDNSKNTAYCQVGGRNLHDR LRLYEALVANRTFVVYCAVTDGGKTSGQGGQITI SS bzsoo_iusee_4_0_5_9_6_4_20_5_21_12_22_5_yf EIQVTVSDVTLSTAELIAKIAAGESIRVSCAVSGTL d0s TSSTVATCVWQAPGGGLLGQSAATSASSSTTLA DSPVFRVTASIDNSKNTVTCTLSAASTPRTAQNA VLRWAARSGVLTVTCTITDGSDTSTSTGQLTISS crdfw_lktjp_4_0_5_9_6_4_20_4_21_13_22_4_ll10 EKQVLVSSVTLSLQALWERWAAGQPLRVSCAIS 5 GPLAPSDVVSCQVQAPGKGVLATASAKQGETGT STGDSYGVRVTVSVDGSTNTATCQVSVVTPQAR RRLYRDLLLSGQGLVVTCALISGGKVEAQASGTIT AS crdfw_lktjp_4_0_5_9_6_4_20_4_21_13_22_4_nhu EKQVLVSSVTLSWTEIYQTLLAGQPLRVSCAISGP s3 VAPSDVVSCQVQAPGKGVLATASAKQGETGTST GDSYGVRVTVSVDGSTNTATCQVSATTRQAARA LALSLYRDGKGLVVTCALISGGKVEAQASGTITAS crdfw_lktjp_4_0_5_9_6_4_20_4_21_13_22_4_pqv EKQVLVSSVTLSPAELARRLVGGEPLRVSCAISG dx PVAPSDVVSCQVQAPGKGVLATASAKQGETGTS TSDSYGVRVTVSVDGSTNTVTCQVQATNPRVGR ALWRDLYLSGEGLVVTCALISGGKVEAQGSGTIT AS crdfw_lktjp_4_0_5_9_6_4_20_4_21_13_22_4_t6qo EKQVLVSSVTLSLAEAVQRFAAGEGLRVSCAISG x PVAPSDVVSCQVQAPGKGVLATASAKQGETGTS TSDSYGVRVTVSVDGSTNTATCQVKATTPQAAR RFVRDLVLSGKGLVVTCALISGGKVEAQASGTITA S lmbug_00fye_4_0_5_9_6_4_20_4_21_10_22_5_a EWQVLVSPVTLPQAAILAALAAGEGIRVSCAVSP aocu PVLPSTTVVCVWQAPGGGLLAEGAVSASGTSTE LASSGGYRVTVSFDNSKGTATCVLSGATPNARR GLYRRALLGELVVTCAVLVDGALVAQASATIVVS lmbug_00fye_4_0_5_9_6_4_20_4_21_10_22_5_cz EWQVLVSPVTLPAAELAARLAAGEGIRVSCAVSG sew PVLPSTTVVCVWQAPGGGLLAEGAVSASGTSTE LASSGGYRMTVSFDNSKGTATCVLRGENRSTRN RLYRLAYEGTLVVTCAVLVDGALVAQASATIVVS lmbug_00fye_4_0_5_9_6_4_20_4_21_10_22_5_n EWQVLVSPVTLPAAELTARILAGEGIRVSCAVSG v8cb PVLPSTTVVCVWQAPGGGLLAEGAVSASGTSTE LASSGGYRMTVSFDNSKGTATCVLRAETPNQRR RLIRLAYEGGLVVTCAVLVDGALVAQASATIVVS lmbug_00fye_4_0_5_9_6_4_20_4_21_10_22_5_pr ELQVLVSPVTLPYQEALARIAAGEGIRVSCAVSGP 6id VLPSTTVVCVWQAPGGGLLAEGAVSASGTSTEL ASSGGYRMTVSFDNSKGTATCVLTPENRNAANA LLLKMYRGDLVVTCAVLVDGALVAQASATIVVS lmbug_jdnu4_4_1_5_9_6_4_20_4_21_11_22_5_q EAQVLSVSGVLLTLAELQALLAAGQGIRVSCAVS ainq GPLQPSTLVSCQLQAAGGGLIGSAAAKAGQTGT YYSDSTGVRVTVSVDNSKNTAYCQVGPPNPQGA RALLLRAARYPTFVVYCAVTDGGKTSGQGGQITI SS lmbug_l52uq_4_0_5_9_6_4_20_4_21_12_22_5_ic EIQVLVSPVTLTFAEAAALVAAGQGIRVSCAISGT 3s7 LQPGTRVTCVWQAPGGGLLGQAAVTVGGASTV LADSPAYRVTVSVDNSKNTAYCQLTPANKKAALS LLHRTLASGKGLVVTCTVEDGGDTDTQTGQVTL SS lmbug_lktjp_4_0_5_9_6_4_20_4_21_13_22_4_484 EKQVLVSSVTLSPAALLARLAAGTGLRVSCAISGP 2v VAPSDVVSCQVQAPGKGVLATASAKQGETGTST GDSYGVRVTVSVDGSTNTATCQVSTATPQARRR LARDILLSGQGLVVTCALISGGKVEAQASGTITAS lmbug_lktjp_4_0_5_9_6_4_20_4_21_13_22_4_gd2 EKQVLVSSVTLSRADIVAKILAGEGLRVSCAISGP 2m VAPSDVVSCQVQAPGKGVLATASAKQGETGTST GDSYGVRVTVSVDGSTNTATCQVSASNRNAQR RLVRLALLEPEWLVVTCALISGGKVEAQASGTITA S lmbug_lktjp_4_0_5_9_6_4_20_4_21_13_22_4_jdx EKQVLVSSVTLSLAEVAARLAAGQGLRVSCAISG 9f PVAPSDVVSCQVQAPGKGVLATASAKQGETGTS TGDSYGVRVTVSVDGSTNTATCQVSASNANARR RLWRYIYDSGQGLVVTCALISGGKVEAQASGTIT AS lmbug_lktjp_4_0_5_9_6_4_20_4_21_13_22_4_mo EKQVLVSSVTLSLDEIAARLIAGEGLRVSCAISGPL zgz APSDVVSCQVQHPGKGVLATASAKQGETGTSTG DSYGVRVTVSVDGSTGTATCQVSAGNRQAARKL YLDMYRNPQGLVVTCALISGGKVEGQASGTITAS lmbug_lktjp_4_0_5_9_6_4_20_4_21_13_22_4_ot8 EKQVLVSSVTLSWAQLAAMLAAGQGLRVSCAIS 09 GPVAPSDVVSCQVQHPGKGVLATASAKQGETGT GTGDSYGVRVTVSVDGSTNTATCQVSFVNRQAR NRFLLDTYRSGVGLVVTCALISGGKVEAQASGTIT AS lmbug_lktjp_4_0_5_9_6_4_20_4_21_13_22_4_qsu EKQVLVSSVTLSPAELMARLAAGEGLRVSCAISP 70 PVAPSDVVSCQVQAPGKGVLATASAKQGETGTS TGDSYGVRVTVSVDGSTNTATCQVSALNRQAAR RLVRDLYLTGQPLVVTCALISGGKVEAQASGTITA S lmbug_lktjp_4_0_5_9_6_4_20_4_21_13_22_4_x0 EKQVLVSSVTLSYTDFYQALTSGKGLRVSCAISG wng PVAPSDVVSCQVQAPGKGVLATASAKPGETGTG TNDSYGVRVTVSVDGSTGTATCQVTAANENAAR ALLLRILRDGQGLVVTCALISGGKVEAQASGTITA S lmbug_q50j3_4_0_5_9_6_4_20_5_21_13_22_4_g ELQVTASDVTLTYAEFVQMIVDGKPIRVSCAVSA ydyy PLASLDSVSCVVQAPSGRTLGTGAYNVQTGTTT SADSVGVRVTVSVDNSTNTVTCQAQFTGNRNQR NRLIREAYFSGTGLVVTCTAVAGGTTKTQTAQITI SS lmbug_q50j3_4_0_5_9_6_4_20_5_21_13_22_4_p ELQVTISDVTLTPAQLWEAMSSGKPIRVSCAVSA v5o4 PYASLDSVSCVVQAPSGRTLGTGAYNAQTGTTT SADSVGVRVTVSVDNSTNTVTCQATFNASRNTIN RIVREVALTGQGLVVTCTAVAGGTTKTQTAQITIS S lmbug_uqwk8_4_0_5_9_6_4_20_4_21_9_22_4_ve EIQVLVSDLTLPLAEVVAKIRAGEPLRVSCAISSD q77 GTTVDEVSCVWQAAGGGLLGEAAVDASGNTTY YAGSPAFRVTVSYDNSKGTAYCQLSAANKRQAIS LFQRYNSLVVTCAAYSGGKVSGQGTGTVTLS u3vsl_pferk_4_0_5_9_6_5_20_5_21_11_22_4_52j EIQVLLSDRTLTPEELRQEIQANGGTLRVSCAVS bz GPLPASATVSCVIQWAGGGSELVGTANVGTGSS TTADSGAWRGTASVDNSKNTVTCQLTAYNAKHA TALASRLVILGRVVFTCAVIVNGELKGQGSGTLTI S u3vsl_pferk_4_0_5_9_6_5_20_5_21_11_22_4_67 EIQVLLSDRTLTRAQLLAEMAANGNVLRVSCAVS n41 GPLPASATVSCVIQWAGGGSELVGTANVGTGSS TTADSGAWRGTASVDNSKNTVTCQLEVRSEKDA VRFADLLYAKGKVVFTCAVIVNGELKGQGSGTLTI S u3vsl_pferk_4_0_5_9_6_5_20_5_21_11_22_4_9ta EIQVLLSDRTLTPAQLLAEMTANGGTLRVSCAVS qg GPLPASATVSCVIQWAGGGSELVGTANVGTGSS TTADSGAWRGTASVDNSKNTVTCQLTARNARA MTALVNRWALLGKVVFTCAVIVNGELKGQGSGT LTIS Example 5 – Production and characterization Polypeptides provided herein can be produced by expression in CHO cells. Purification can be performed with His-tag purification, and quality control can be done with A280, SDS- PAGE / Caliper QC. The produced polypeptides can then be characterized for binding and gastrointestinal stability. Briefly, binding characterization can be performed by biolayer interferometry (BLI), which can include screening at single analyte concentration to identify binders and full characterization of binder affinity and kinetics. Gastrointestinal stability can be performed by incubation in simulated gastric fluid (SGF) and simulated intestinal fluid (SIF) at digestion- relevant timescales and beyond (e.g., 0hr, 4hr, 24hr). SDS-PAGE / Caliper can be used to assess degradation. Example 6 – Single domain antibodies with helical framework loops A number of exemplary single domain antibodies with helical framework loops were generated. Antibody design The single domain antibodies with helical framework loops were designed by first generating a protein backbone, then generating an amino acid sequence for that backbone. The resulting designs were evaluated with computational metrics. The steps taken were as follows. 1. Generation of backbones for antibodies with helical framework loops: a. Backbones were computationally generated using a denoising diffusion probabilistic model (hereafter ‘diffusion model’), which is a generative model that outputs backbone atom coordinates for a protein. This model is used with fold conditioning, a method which allows for the specification of a protein fold. The fold was specified by providing each secondary structural element (alpha helix, beta sheet, or loop), its length in number of residues, and its contacts with other secondary structural elements. For the exemplary single domain antibodies with helical framework loops, a single domain antibody fold was specified with loop 1 of framework 1 and loop 3 of framework 3 replaced by alpha helices. b. Backbones were designed to bind to a target by providing the model with a target structure and desired residues on that structure (i.e., the epitope). Backbones for these examples were designed using the docking approach, followed by partial diffusion: i. Docking approach: A backbone for an antibody with helical framework loops was generated without providing a target, using the diffusion model with fold conditioning to specify the fold. Then, a backbone consisting of two or three helices in a helical bundle fold was generated to bind to a target, using the diffusion model with fold conditioning as well as a target structure and target residues on that structure. Two of these helices were then used to dock the antibody with helical framework loops backbone. ii. Partial diffusion: The diffusion model was used in a ‘partial diffusion’ mode to redesign a designed antibody with helical framework loops. This method involved adding some noise to the backbone coordinates, then using the diffusion model to generate a new backbone from the ‘noisy’ backbone. This method can provide additional structural variation and yield improved designs. c. The result was a designed backbone for an antibody with helical framework loops in complex with a target structure, docked at specified target residues on that structure and refined using partial diffusion. 2. Generation of sequences: a. Amino acid sequences were generated for the backbones using a message- passing neural network model. The sequences for the target and the antibody scaffold were provided to the model along with the backbone structure of the complex (consisting of target and antibody with helical framework loops), and the sequence was generated for the helical framework loops in this context. If necessary for improving predicted accuracy and confidence (see ‘Evaluation of computational designs’), some of the amino acids of the antibody scaffold sequence can be redesigned to better support the helical framework loops. In particular, the amino acids neighboring the helical framework loops can be redesigned. 3. Evaluation of computational designs: a. Designs were evaluated using protein structure prediction models. These models provided a predicted structure given the sequence of the design, which was aligned to the design structure model to assess accuracy. These models can also provide a measure of confidence in the prediction. Designs that are predicted with higher accuracy and higher confidence are more likely to function experimentally as designed. Production of example single domain antibodies with helical framework loops Exemplary single domain antibodies with helical framework loops were produced using either transient Chinese hamster ovary (CHO) cell expression or E. coli SHuffle expression, followed by His-tag purification. The steps for CHO production were as follows. 1. Plasmid cloning: a. Synthesized gene fragments were ordered (eBlock gene fragments, IDT) consisting of codon-optimized sequences for designed single domain antibodies with helical framework loops, including 6xHis tag and secretion signal. b. Gene fragments were cloned into a vector using Golden Gate assembly. c. Golden Gate assembly product was transformed into competent TOP10 E. coli (Thermo Fisher) and grown overnight in 1mL TB, then plasmids were purified (ZymoPURE 96 Plasmid Miniprep Kit) from overnight culture. 2. CHO cell transfection and expression: a. Purified plasmid was used to transfect CHO cells using the ExpiCHO Expression System Kit (Thermo Fisher) according to the protocol in 96 well (1mL) format. b. Single domain antibodies with helical framework loops were expressed in CHO cells according to the protocol (ExpiCHO Expression System Kit, Thermo Fisher). c. CHO supernatant containing single domain antibodies with helical framework loops was harvested following expression. The steps for E. coli SHuffle production were as follows. 1. Plasmid cloning: a. Synthesized gene fragments were ordered (eBlock gene fragments, IDT) consisting of codon-optimized sequences for designed single domain antibodies with helical framework loops, including 6xHis tag. b. Gene fragments were cloned into a vector using Golden Gate assembly. 2. E. coli SHuffle expression: a. Golden Gate assembly product was transformed into SHuffle T7 Competent E. coli (New England Biolabs). b. Single domain antibodies with helical framework loops were expressed in E. coli SHuffle. c. E. coli SHuffle pellets were harvested and lysed following expression. His-tag purification 1. Single domain antibodies with helical framework loops were purified from CHO expression supernatant or E. coli SHuffle cell lysate using His-tag immobilized metal affinity chromatography (IMAC) in 96 well format using either nickel-nitrilotriacetic acid resin columns or magnetic agarose beads (HisPur Ni-NTA Spin Plates, Thermo Fisher or Ni-NTA Magnetic Agarose Beads, Pierce). Simulated gastric fluid (SGF) and simulated intestinal fluid (SIF) stability assay protocol 1. Aliquot 7.5 uL of His-tag purified protein sample in elution buffer (25 mM NaPhosphate pH 7.4, 275 mM NaCl, 250 mM Imidazole). 2. Add 7.5 uL of 2x freshly prepared SGF or SIF solution: a. SGF solution prepared per the US Pharmacopeia standard: 600 ug / mL pepsin and 34.2 mM NaCl in water, with HCl added to adjust pH to 2 (all at 1x, adjusted for 2x). b. SIF solution (see table below for recipe; buffer mixes sourced from BioRelevant) with added proteases trypsin at 60 μg / mL and chymotrypsin at 60 μg / mL (yielding 30 μg / mL each at 1x). 3. Incubate at 37^C for 4 hour timepoint. For 0 hour timepoint, add elution buffer to match the volume of the digestion timepoints. 4. Add 5.5uL of 4x Tris / Glycine / SDS running buffer (BioRad) and 1uL 1M DTT to each sample and heat 5 minutes at 95^C to end SIF incubation. 5. Run samples immediately on 4-20% TGX gel (BioRad) to perform SDS-PAGE analysis using Precision Plus Protein Dual Xtra Prestained Protein Standards (BioRad) as ladder. Table 2 Table 3 SGF and SIF stability SDS-PAGE analysis Samples were expected to run at around 15 kDa given their molecular weight and observed at approximately 15kDa. Protease bands were also visible on the gel. For the designed single domain antibodies with helical framework loops, intact sample was visible on the gel as indicated by the arrows at both the 0h and 4h timepoints. These results indicate that the designed single domain antibodies with helical framework loops survived for at least 4 hours of incubation in both SGF and SIF. Biolayer interferometry (BLI) assay and analysis Binding was assessed by loading biotinylated target onto streptavidin coated biosensor. Association and dissociation were measured using His-tag purified sample, with background subtracted from a blank sample to remove buffer effects. Affinity was estimated by fitting curves to obtain kinetic parameters. Nano differential scanning fluorimetry (DSF) assay and analysis Melting curves were generated using nano DSF (Uncle, WGQ-DPD-0061, Unchained labs, Uncle Client V5.04) with a start temperature of 25℃, a scanning rate of 0.4 ℃ / min, and a final temperature of 95℃. The first derivative of the signal was used to calculate the inflection point to obtain a value for melting temperature. Minor inflection points were ignored as these do not represent the main unfolding transition. Size exclusion chromatography (SEC) assay and analysis Size exclusion chromatography was performed using a standard setup and protocol (Agilent) to assess monodispersity and aggregation of single domain antibodies with helical framework loops. Monomeric protein was expected and observed to elute at 10-11min given the molecular weights, approximately 15 kDa for the samples. The elution times and clean peaks indicate that the designed single domain antibodies with helical framework loops are monodisperse and do not aggregate. An exemplary antibody was produced that target TNFα. The antibody was designed as provided above and produced by CHO expression. The sequence for the exemplary antibody is shown below. A BLI binding curve and nano DSF melting temperature curve for the designed single domain antibody with helical framework loops are shown in FIGS.6 and 7. The antibody was expected to exhibit low to sub nanomolar KD with a melting temperature Tm of 92.1^C. hFL_TNFa_1, helical framework loops in bold: EIQVTVSDVTLSTAELIAKIAAGESIRVSCAVSGTLTSSTVATCVWQAPGGGLLGQSAATSAS S STTLADSPVFRVTASIDNSKNTVTCTLSAASTPRTAQNAVLRWAARSGVLTVTCTITDGSDTS T STGQLTISS Exemplary antibodies were produced that target IL7Rα. The antibodies were designed as provided above and produced by CHO expression. The sequences for the exemplary antibodies are shown below. BLI binding curves and nano DSF melting temperature curves for the designed single domain antibodies with helical framework loops are shown in FIGS.8-11. The antibodies were expected to exhibit mid or high nanomolar KDwith a melting temperature Tm of greater than 95^C and 82^C, respectively. hFL_IL7Ra_1, helical framework loops in bold: ETQTLTCTDVLLSEEEFEEMRLKAIKGEPITIRVSCAVSGEFLPSDEISCGVQLPGKGLEWVS A YNVGTGETYYADSGGLRATVSVDNSKNTVYCQLTVEGKEEQEEILKLLAKGEGVVVTCAL LRDG KLVGQGTAQIKLS hFL_IL7Ra_2, helical framework loops in bold: ETQTLTSTDVLLSEEEFEEMLSKLLNGEPVSIRVSCAVSGEFLPSDEISCGVQLPGKGLEWVS A YNVGTGETYYADSSGLRATVSVDNSKNTVYCQLSGKTREAWEEFLKTLAEGKSVVVTCAL LRDG KLVGQGTAQIKLS Exemplary antibodies were produced that target IL23α. The antibodies were designed as provided above and produced by CHO expression or E. coli SHuffle production. The sequences for the exemplary antibodies are shown below. BLI binding curves and nano DSF melting temperature curves for the designed single domain antibodies with helical framework loops are shown in FIGS. 12-31. The antibodies were expected to exhibit mid micromolar or high nanomolar KD with a melting temperature Tm of 83.9^C, 86.0^C, 91.1^C, >95^C, >95^C, 70.3^C, >95^C, 77.5^C, 72.1^C, or 84.1^C, respectively. hFL_IL23a_1, helical framework loops in bold: EIQVLVSPVTLTMDEVMALALAGEGIRVSCAISGTLQPGTRVTCVWQAPGGGLLGQAAVTV GGASTVLADSPKYRVTVSVDNSKNTAYCQLSGREPWYLLALADQALRSGKGLVVTCTVED GGDTDTQ TGQVTLSS hFL_IL23a_2, helical framework loops in bold: EIQVLVSPVTLTMDEVLALIQAGEPIRVSCAISGTLQPGTRVTCVWQAPGGGLLGQAAVTVG GASTVLADSPKYRVTVSVDNSKNTAYCQLSTTEPVLMLALAHRALRSGRGLVVTCTVEDG GDTDTA TGQVTFSS hFL_IL23a_3, helical framework loops in bold: EIQVLVSPVTLTMDEVLALALAGEGIRVSCAISGTLQPGTRVTCVWQAPGGGLLGQAAVTGG GASTVLADSPKYRVTVSVDNSKNTAYCQLLGERPIYLLALAHQAVASGRGLVVTCTVEDGG DTDTQ TGQVTVSS hFL_IL23a_4, helical framework loops in bold: EIQVLVSPVTLTMEEVLALLAAGQGIRVSCAISGTLQPGTRVTCVWQAPGGGLLGQAAVTV GGASTVLADSPKYRVTVSVDNSKNTAYCQLHAAEPWEMLALADEALRSGKGLVVTCTVE DGGDTDTQ TGQVTVSS hFL_IL23a_5, helical framework loops in bold: EIQVLVSPVTLTMDEVMALALAGEPIRVSCAISGTLQPGTRVTCVWQAPGGGLLGQAAVTV GGASTVLADSPKYRVTVSVDNSKNTAYCQLSSREPYALLAEADRALRSGRGLVVTCTVEDG GDTGTQ TGQVTLSS hFL_IL23a_6, helical framework loops in bold: EIQVLVSPVTLTMDEVMALALAGEPIRVSCAISGTLQPGTRVTCVWQAPGGGLLGQAAVTV GGASTVLADSPKYRVTVSVDNSKNTAYCQLVAARPEDLLALAYLALRSGKGLVVTCTVED GGDTDTQ TGQVTLSS hFL_IL23a_7, helical framework loops in bold: ETQTLTSTDVLLSEEQLVALWERTVRGEPATIRVSCAVSGELLPSDEISCGVQLPGKGLEWV SAYNVGTGETYYADSSGLRATVSVDNSKNTVYCQLIITDARDMLRLILLLISGQNVVVTCAL LRDG KLVGQGTAQIKLS hFL_IL23a_8, helical framework loops in bold: ETQTLTSTDVLLTEQEVADILLAAFRGEPATIRVSCAVSGEFLPSDEISCGVQLPGKGLEWVS AYNVGTGETYYADSSGLRATVSVDNSKNTVYCQLTIEDPELALRFALALIRGEPVVVTCALL RDG KLVGQGTAQIKLS hFL_IL23a_9, helical framework loops in bold: ELQVSVSDVTLSQTDVVNEWLANGGSVRFSCAISGPLSPDAVVSCVLQDAGGGLEASVAFK VSESSASAAGSWGGRLTVSVDASKGTVTCVVSAANLAQFQLFMEALLAGLVLTCTASSNG DTATASGQVTASS hFL_IL23a_10, helical framework loops in bold: EIQVLVSPVTLTPQEVLRLIASGEPIRVSCAISGTLQPGTRVTCVWQAPGGGLLGQAAVTVGG ASTVLADSPKYRVTVSVDNSKGTAYCQLSAARGEALLAELARGYLSGRGLVVTCTVEDGG DTDTQ TGQVTLSS Simulated gastric and simulated intestinal fluid incubation experiments and size exclusion chromatography traces for designed single domain antibodies with helical framework loops (targeting TNFa, IL23a, or TL1A) were performed. The example antibodies were designed as provided above and produced by CHO expression or E. coli SHuffle production. The results from the simulated gastric and simulated intestinal fluid incubation experiments using the single domain antibodies are shown below (FIGS.32-37). The results show an estimated survival time of at least 4 hours in simulated gastric fluid (SGF) and simulated intestinal fluid (SIF). Size exclusion chromatography traces of certain example antibody samples show the samples were monodisperse and do not aggregate (FIGS.38 and 39). Predicted structures of certain of the designed single domain antibodies with helical framework loops are shown in FIGS.40-45. hFL_TNFa_2, helical framework loops in bold: EVQVTVSDVTLTLEDLAALLDAGEGVRQSCAVSTSSDVPVTVTCVVQAPGGGLLGSATVSS SES STTGAGSPGWRVTVSYDNSKGTATCVLEAATRTARNRLLLTAYRTGVAATFTCTATAGSDT ATATGQVTVSS hFL_TNFa_3, helical framework loops in bold: EVQVTVSDVTLTEAELWARLAAGEGVRQSCAVSTSSDVPVTVTCVVQAPGGGLLGSATVSS SESSTTGAGSPGWRVTVSYDNSKGTATCVLSALTPNAARRLWTESVLSGRGPTFTCTATAGS DTATATGQVTVSS hFL_TNFa_4, helical framework loops in bold: EVQVTVSDVTLTVDDLIAMLNAGEGVRVSCAVSTSSDVPVTVTCQVQAPGGGLLGSATVSS SES STTGAGSGGWRVTVSYDNSSGTATCVLRPLTENARRALMKNAMITGKLATFTCTATAGSDT ATATGQVTCSQ hFL_IL23a_11, helical framework loops in bold: ELQVTVSDVTLSKEEAIFLLYEGKGLRSSCAVSTSSPAPVTVSCVVQAPSGELIGRGTLNSGT G QTYSDTSLKGRVTLSYDNSKNTVYCQLSCETEEKRQEVMTLLYSSDSLVVTCTASQGGETA TDTGQITLSS hFL_IL23a_12, helical framework loops in bold: ELQVLVSDVTLTLEEAGEMLLRGEPLRSSCAVSGELQPSDVVSCSWQAAGGGLLGSAAASA SESSTTAAGSAGFRVTVSIDNSKNTVTCSLGPADELARVRLLALLERERTLVVTCAVLRNGE LVGQASGRITLS hFL_IL23a_13, helical framework loops in bold: EVQVLASDVTLTLEELLARAEAGDLSVRVSCAVSAPKGLPTTVSCYLQLPGKGLELLATANL AT GETTSHDSGGFRGTLSVDQSKGTVTCTVTPVDEAAFAQAVLTGLKTGTLFVVTCVAESGGE TATATAQVTVSS hFL_TL1A_1, helical framework loops in bold: EVQVLVSDVTLPKEEAIKRIEEGEPLRVSCAVSSSSSSPVTVSCQLQAADGTLLGSGTLNTGT G ATTSGDSSLVRVTLSYDNSTGTVTCQFSPIDDLAREELIDLIQEQDTLVVTCTASGGSQTATA T GQITFSS hFL_TL1A_2, helical framework loops in bold: ELQVTVSDLTLSFEEVEELLKAGEGLRVSCAVSRPLSDSDVVSVVWQTLGGGLLASVAGKA SEGSTTGVDSSGVRVTLSVDKSKGTVTSQLRCETEEAREFLLSELELEGALVVYCTVKDGSQ TATAT GQVTFSS hFL_TNFa_5, helical framework loops in bold: ELQVTGSDVTLSVSEVLSMIAAGEPLRVSCAVSGPLSPSDVVSAVVQAVDGTLLGSVAAKAS ES STTGVDSSGIRVTLSIDKSKNTVYVQLGPANRNGANLLARLAYEQKQLVVTCALKRGGQTS TGT AQVTFSS hFL_TNFa_6, helical framework loops in bold: ELQVTVSDVTLTAAEFVQLLNSGEGVRQSCAVSTSSDVPVTVTCVVQAPGGGLLGSATVSSS ES STTAAGSPGWRVTVSYDNSKGTVTCVLQALTENERRRLIREALITGVGPTFTCTATAGSDTA TA TAQVTLSS hFL_TNFa_7, helical framework loops in bold: EVQVTVSDVTLTEAEALRLLLSGEPVRVSCAVSTSSDVPVTVTCVVQAPGGGLLGSATVSSS EG STTVASSPDWEVTVSYDNSKGTATCGLRVTNRFSAIKLYDEAAASGVLATFTCTATGGSDTA TA TGQVTLSS References: 1. Jumper, J., Evans, R., Pritzel, A. et al. Highly accurate protein structure prediction with AlphaFold. Nature 596, 583–589 (2021). https: / / doi.org / 10.1038 / s41586-021-03819-2 2. Minkyung Baek et al., Accurate prediction of protein structures and interactions using a three-track neural network. Science 373, 871-876(2021). DOI:10.1126 / science.abj8754 3. Ruffolo, J.A., Chu, LS., Mahajan, S.P. et al. Fast, accurate antibody structure prediction from deep learning on massive set of natural antibodies. Nat Commun 14, 2389 (2023). https: / / doi.org / 10.1038 / s41467-023-38063-x 4. Cao, L., Coventry, B., Goreshnik, I. et al. Design of protein-binding proteins from the target structure alone. Nature 605, 551–560 (2022). https: / / doi.org / 10.1038 / s41586-022-04654-9 5. Watson, J.L., Juergens, D., Bennett, N.R. et al. De novo design of protein structure and function with RFdiffusion. Nature 620, 1089–1100 (2023). https: / / doi.org / 10.1038 / s41586-023- 6. Vázquez Torres, S., Leung, P.J.Y., Venkatesh, P. et al. De novo design of high-affinity binders of bioactive helical peptides. Nature 626, 435–442 (2024). https: / / doi.org / 10.1038 / s41586- 7. Bennett, N.R., Coventry, B., Goreshnik, I. et al. Improving de novo protein binder design with deep learning. Nat Commun 14, 2625 (2023). https: / / doi.org / 10.1038 / s41467-023-38328-5 8. Svilenov, H.L., Sacherl, J., Protzer, U. et al. Mechanistic principles of an ultra-long bovine CDR reveal strategies for antibody design. Nat Commun 12, 6737 (2021). https: / / doi.org / 10.1038 / s41467-021-27103-z 9. Kadonosono, T., Yimchuen, W., Ota, Y. et al. Design Strategy to Create Antibody Mimetics Harbouring Immobilised Complementarity Determining Region Peptides for Practical Use.Sci Rep 10, 891 (2020). https: / / doi.org / 10.1038 / s41598-020-57713-4 10. Abskharon R, Pan H, Sawaya MR, et al. Structure-based design of nanobodies that inhibit seeding of Alzheimer's patient-extracted tau fibrils. Proc Natl Acad Sci U S A. 2023;120(41):e2300258120. https: / / doi.org / 10.1073 / pnas.2300258120 11. Uchański, T., Masiulis, S., Fischer, B. et al. Megabodies expand the nanobody toolkit for protein structure determination by single-particle cryo-EM. Nat Methods 18, 60–68 (2021).

Claims

CLAIMS What is claimed is:

1. A polypeptide comprising at least one helical framework loop, optionally wherein the at least one helical framework loop comprises an alpha helix secondary structure, optionally, wherein the polypeptide binds a target, is resistant to degradation in acidic conditions, such as at a pH of the stomach, is resistant to degradation at an intestinal pH, is thermostable, is shelf stable, is orally deliverable and / or is resistant to protease or proteolytic degradation, further optionally, wherein the alpha helix comprises the sequences of any one of the alpha helixes provided herein. 2 The polypeptide of claim 1, wherein the polypeptide comprises at least two helical framework loops, wherein each of the least two helical framework loops comprises an alpha helix secondary structure. 3 The polypeptide of claim 2, wherein the polypeptide comprises at least three helical framework loops, wherein each of the at least three helical framework loops comprises an alpha helix secondary structure. 4 The polypeptide of any one of claims 1-3, wherein the polypeptide further comprises an antibody scaffold. 5 The polypeptide of claim 4, wherein the antibody scaffold is a single-domain antibody (sdAb), single-chain variable fragment (scFv), Fab’, fragment antigen binding (Fab), F(ab’)2or a full-length antibody without framework loop(s) replaced by the helical framework loop(s). 6 The polypeptide of any of of claims 1-5, wherein the at least one helical framework loop replaces a FR1 loop or FR2 loop or FR3 loop of the antibody scaffold. 7 The polypeptide of claim 2, wherein the at least two helical framework loops replaces a FR1 loop and / or a FR2 loop and / or a FR3 loop of the antibody scaffold.

8. The polypeptide of claim 3, wherein the at least three helical framework loops replaces a FR1 loop, a FR2 loop and a FR3 loop of the antibody scaffold.

9. The polypeptide of any one of claims 1-8, wherein each helical framework loop is 5-20 residues in length.

10. The polypeptide of claim 9, wherein each helical framework loop is 5-14 residues in length.

11. The polypeptide of any one of claims 1-10, wherein the polypeptide further comprises a linker at the N-terminus and / or C-terminus of each helical framework loop.

12. The polypeptide of claim 11, wherein the linker does not form an alpha helix or beta sheet secondary structure.

13. The polypeptide of claim 11 or 12, wherein the linker at the N-terminus is 1-8 residues in length.

14. The polypeptide of claim 13, wherein the linker at the N-terminus is 1-5 residues in length.

15. The polypeptide of claim 11 or 12, wherein the linker at the C-terminus is 1-8 residues in length.

16. The polypeptide of claim 15, wherein the linker at the C-terminus is 1-5 residues in length.

17. The polypeptide of any one of claims 1-16, wherein the polypeptide has a binding affinity of at least 10-6M or 10-7M.

18. The polypeptide of any one of claims 1-17, wherein the polypeptide has a binding affinity of at least 10-8M or 10-9M.

19. The polypeptide of any one of claims 1-18, wherein the polypeptide is resistant to degradation at an intestinal pH.

20. The polypeptide of claim 19, wherein the intestinal pH is 6.5.

21. The polypeptide of any one of claims 1-20, wherein the polypeptide is resistant to degradation at a gastric pH.

22. The polypeptide of claim 21, wherien the gastric pH is 2.

23. The polypeptide of any one of claims 1-22, wherein the polypeptide is resistant to degradation at 37ºC or at any one of the temperatures or temperature ranges provided herein.

24. The polypeptide of any one of claims 1-23, wherein the polypeptide is resistant to degradation by one or more proteases.

25. The polypeptide of any one of claims 1-24, wherein the one or more proteases comprises trypsin and / or chymotrypsin.

26. The polypeptide of any one of claims 1-25, wherein the one or more proteases comprises pepsin.

27. The polypeptide of any one of claims 1-26, wherein the polypeptide is resistant to degradation in simulated intestinal fluid, such as that of Table 3.

28. The polypeptide of claim 27, wherein the polypeptide is resistant to degradation in simulated intestinal fluid for at least 1 hour, 2 hours, 3 hours or 4 hours.

29. The polypeptide of any one of claims 1-28, wherein the polypeptide is resistant to degradation in simulated gastric fluid, such as that of Table 2.

30. The polypeptide of claim 29, wherein the polypeptide is resistant to degradation in simulated gastric fluid for at least 1 hour, 2 hours, 3 hours or 4 hours.

31. The polypeptide of any one of the preceding claims, wherein the target is IL7Rα, IL23α, TL1A or TNFα.

32. The polypeptide of any one of the preceding claims, wherein the polypeptide comprises a sequence of any one of the sequences provided herein.

33. A composition comprising a polypeptide of any one of the preceding claims, further comprising a pharmaceutically acceptable carrier.

34. A method for producing a polypeptide comprising at least one helical framework loop that binds a target, comprising: a) computationally generating a backbone comprising at least one alpha helix and an antibody scaffold structure, b) generating an amino acid sequence, optionally such that the at least one helical framework loop binds to the target, and c) optionally, evaluating the produced polypeptide using a protein structure prediction model.

35. The method of claim 34, wherein a) comprises using a diffusion model with fold conditioning, wherein the fold is specified with secondary structural elements, which secondary structural elements comprise the at least one alpha helix and one or more structural elements of the antibody scaffold.

36. The method of claim 35, wherein a) comprises i) generating a backbone comprising an antibody scaffold structure using a diffusion model with fold conditioning, wherein the fold is specified with secondary structural elements, without providing a target, ii) generating a backbone comprising at least one, two or three alpha helices that bind to a target using a diffusion model with fold conditioning as well as a target structure and target residues on the target structure, and iii) grafting two of the helices of ii) onto the backbone of i).

37. The method of claim 36, wherein b) occurs before or after iii).

38. The method of any one of claims 34-37, wherein a) comprises providing a target structure and target residues on the target structure using a diffusion model with fold conditioning, wherein the fold is specified with secondary structural elements, which secondary structural elements comprise the at least one alpha helix and one or more structural elements of the antibody scaffold.

39. The method of any one of claims 34-38, wherein b) comprises using a message-passing neural nework model.

40. The method of any one of claims 34-39, further comprising assessing, modeling or selecting for resistance to degradation at a specific pH, such as in acidic conditions, such as at a pH of the stomach, or as in the pH of the intestine, thermostability, shelf stability, oral deliverability and / or resistance to protease or proteolytic degradation.

41. The method of any one of claims 34-40, further comprising assessing, modeling or selecting for resistance to degradation at an intestinal pH, such as a pH of 6.

5.

42. The method of any one of claims 34-41, further comprising assessing, modeling or selecting for resistance to degradation simulated intestinal fluid, such as that of Table 3.

43. The method of any one of claims 34-42, further comprising assessing, modeling or selecting for resistance to degradation at a gastric pH, such as a pH of 2.

44. The method of any one of claims 34-43, further comprising assessing, modeling or selecting for resistance to degradation simulated gastric fluid, such as that of Table 2.

45. The method of any one of claims 34-44, further comprising assessing, modeling or selecting for resistance to degradation at 37ºC or any one of the temperatures provided herein.

46. The method of any one of claims 34-45, further comprising assessing, modeling or selecting for resistance to degradation by one or more proteases.

47. The method of claim 46, wherein the one or more proteases comprises trypsin and / or chymotrypsin.

48. The method of claim 46, wherein the one or more proteases comprises pepsin.

Images