Methods and systems for assembly of nucleic acids

A polynucleotide library with complementary fragment and index regions, combined with AI algorithms, addresses scalability and complexity issues in nucleic acid assembly, enabling efficient construction of large and complex nucleic acids for advanced biological applications.

WO2026236160A1PCT designated stage Publication Date: 2026-11-12GENYRO INC
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
PCT/US2026/026611
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2025-05-05
Filing Date
2026-05-04
Publication Date
2026-11-12

AI Technical Summary

Technical Problem

Existing methods for nucleic acid synthesis and assembly are limited by scalability, automation, speed, accuracy, and cost, particularly when constructing large sequences or those with high complexity, such as repetitive sequences or GC-rich DNA, hindering advancements in 'big biology' and gene design.

Method used

A polynucleotide library design with complementary fragment and index regions, using nick repair enzymes and ligases, along with AI-informed algorithms for fragment assembly, enabling efficient construction of complex nucleic acids up to genome scale.

Benefits of technology

Facilitates rapid assembly of large, complex nucleic acids with high fidelity and scalability, allowing for the creation of synthetic proteins, genomes, and gene circuits, and supports AI-informed design of biological entities.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure US2026026611_12112026_PF_FP_ABST
    Figure US2026026611_12112026_PF_FP_ABST
Patent Text Reader

Abstract

Provided herein are methods, systems, and compositions for efficient nucleic acid assembly. Nucleic acid assembly may comprise assembly of variants comprising paired homology.
Need to check novelty before this filing date? Find Prior Art

Description

WSGR Docket No.: 69686-701.601METHODS AND SYSTEMS FOR ASSEMBLY OF NUCLEIC ACIDSCROSS-REFERENCE

[0001] This application claims the benefit of U.S. Provisional Patent Application No.63 / 800,237 filed on May 5, 2025, the disclosure of which is incorporated by reference in its entirety.BACKGROUND

[0002] De novo nucleic acid synthesis is a powerful tool for biological research and biotechnology applications. While various methods are known for the synthesis and assembly of relatively short fragments of nucleic acids on a small scale, these techniques suffer from lack of scalability (the ability to construct large, up to genome-sized DNA fragments), as well as insufficient automation, speed, accuracy, and high cost when applied to larger sequences (including entire genomes). Such methods are, unable to efficiently construct sequences of high complexity (for example, those containing repetitive sequences, and / or that are GC rich - for example, at least 50% of the human genome is known to comprise such complex DNA and 85% of plant genomes). Thereby placing a significant constraint on the type of DNA sequences that can be constructed, and making it challenging to explore the possibilities of ‘big biology’ and to fully leverage new Al-informed gene and whole genome design capabilities using foundational models to discern the ‘generative rules of biology’ to transform DNA into a predictive engineering material. Thus, a significant need remains for methods of nucleic acid assembly and construction able to address these issues.BRIEF SUMMARY

[0003] Provided herein are compositions, methods, and systems for nucleic acid assembly.

[0004] Provided herein are compositions comprising a polynucleotide library, wherein the polynucleotide library comprises: (a) a first polynucleotide, wherein the first polynucleotide comprises a first strand and a second strand, the first strand comprises a first fragment region, a first stabilizing region, and a first index region; the second strand comprises a second fragment region, wherein the first fragment region and the second fragment region are at least partially complementary; and (b) a second polynucleotide, wherein the second polynucleotide comprises a third strand and a fourth strand, wherein the third strand comprises a second index region and a third fragment region, and the fourth strand comprises a second stabilizing region and a fourth fragment region; wherein the third fragment region and the fourth fragment region are at least partially complementary; and wherein the first index region andWSGR Docket No.: 69686-701.601the second index region are at least partially complementary, and the first stabilizing region and the second stabilizing region are at least partially complementary; wherein the polynucleotides of the library collectively comprise one or more complex regions, and the polynucleotide library collectively encodes for a synthetic protein, aptamer, synthetic genome, synthetic gene circuit, or synthetic vector. Further provided herein are compositions wherein the one or more complex regions individually or collectively comprise secondary structure (optionally alpha helix, beta sheet, stem-loop, pseudoknot, homodimer, heterodimer, hairpin loop, interior loop, bulge, or multiloop), inversions, palindromic sequences, repeats, high GC, or low AT. Further provided herein are compositions wherein the one or more of the fragment regions individually or collectively comprise at least two complex regions. Further provided herein are compositions wherein one or more of the first fragment region, the second fragment region, the third fragment region, and the fourth fragment region is independently 100-1000, 100-2500, 500-5000, 500-10,000, or 500-100,000 nucleotides in length. Further provided herein are compositions wherein one or more of the index sequences is independently 10-200 nucleotides in length. Further provided herein are compositions wherein one or more of the stabilizing sequences is independently 10-200 nucleotides in length. Further provided herein are compositions wherein the first stabilizing region and the second stabilizing region have a Tmof 40-95 degrees Celsius. Further provided herein are compositions wherein the composition further comprises a nick repair enzyme. Further provided herein are compositions wherein the composition further comprises a ligase. Further provided herein are compositions wherein the composition further comprises a buffer. Further provided herein are compositions wherein the composition further comprises a buffer having a pH of 5-9. Further provided herein are compositions wherein the first index region and the second index region have a Tmof 40-95 degrees Celsius. Further provided herein are compositions wherein the first index region and the second index region are less than 90, 95%, 97%, or 99% identical or complementary to any fragment region in the polynucleotide library. Further provided herein are compositions wherein the first index region and the second index region are different than any fragment region in the polynucleotide library. Further provided herein are compositions wherein any index region in the library comprises a Hamming or Levenshtein distance of at least two from any other index region. Further provided herein are compositions wherein the second stabilizing region and the second fragment region are covalently attached. Further provided herein are compositions wherein the first strand comprises from 5’ to 3’: the first fragment region, the first stabilizing region, and the first index region. Further provided herein are compositions wherein the third strandWSGR Docket No.: 69686-701.601comprises from 5’ to 3’: the second index region and the third fragment region. Further provided herein are compositions wherein the fourth strand comprises from 5’ to 3’: the fourth fragment region and the second stabilizing region. Further provided herein are compositions wherein the first strand further comprises a 5’ third index sequence. Further provided herein are compositions wherein the second strand further comprises a third 5’ stabilizing sequence. Further provided herein are compositions wherein the first strand comprises a 3’ overhang. Further provided herein are compositions wherein the 3’ overhang comprises the first stabilizing region and the first index region. Further provided herein are compositions wherein the third strand comprises a 5’ overhang. Further provided herein are compositions wherein the 5’ overhang comprises the second index region. Further provided herein are compositions wherein the second nucleic acid comprises at least one non-complementary region. Further provided herein are compositions wherein the at least one non-complementary region is formed between the second index sequence and the second stabilizing region. Further provided herein are compositions wherein the one or more of the fragment regions individually or collectively comprise at least 1, 2, 5, 10, 20, 30, 40, 50, 60, 75, 80, or at least 90% complex regions. Further provided herein are compositions wherein the composition does not comprise one or more of a polymerase, exonuclease, endonuclease, SSB, or recombinase. Further provided herein are compositions wherein the 5’ end of the first fragment region and the 3’ end of the third fragment region are not substantially complementary. Further provided herein are compositions wherein the library comprises at least 5, 10, 15, 20, 30, 40, 50, 75, 100, or at least 500 unique polynucleotides.

[0005] Provided herein are kits comprising a composition described herein.

[0006] Provided herein are methods of nucleic acid assembly comprising: assembling two or more polynucleotide fragments into the nucleic acid, wherein the nucleic comprises one or more of complex regions. Further provided herein are methods wherein the one or more complex regions individually or collectively comprise secondary structure (optionally alpha helix, beta sheet, stem-loop, pseudoknot, homodimer, heterodimer, hairpin loop, interior loop, bulge, or multiloop), inversions, palindromic sequences, repeats, high GC, or low AT.Further provided herein are methods wherein the nucleic acid comprises at least two complex regions. Further provided herein are methods wherein the nucleic acid comprises at least 1, 2, 5, 10, 20, 30, or 40% complex regions. Further provided herein are methods wherein the method assembles at least 5, 10, 15, 20, 25, 30, 40, 50 or at least 100 polynucleotide fragments. Further provided herein are methods wherein the nucleic acid is at least 1000, 3000, 5000, 10,000, 20,000, 50,000, 100,000, 500,000, 1 million, 10 million, 100 million, 1WSGR Docket No.: 69686-701.601billion, or at least 10 billion nucleotides in length. Further provided herein are methods wherein the assembling comprises contacting two or more polynucleotide fragments with a nick repair enzyme. Further provided herein are methods wherein the assembling comprises contacting two or more polynucleotide fragments with a ligase. Further provided herein are methods wherein the nucleic acid originates from an animal, plant, bacteria, virus, fungi, or archaea. Further provided herein are methods wherein the nucleic acid originates from a eukaryote. Further provided herein are methods wherein the nucleic acid encodes a protein, a regulatory element, digital information, genome, or vector. Further provided herein are methods wherein the nucleic acid encodes for a protein, wherein the protein optionally comprises collagen, elastin, rubisco, cellulase, fibroblast GF, apolipoprotein, an antibody, or telomerase. Further provided herein are methods wherein the composition collectively encodes for an antibody library having a diversity of at least 106, 107, 108, 109, IO10, 1011or at least 1012unique members. Further provided herein are methods wherein the nucleic acid encodes for a genome, wherein the genome comprises at least 10, 20, 30, 40, 50, 70, or at least 80% repetitive sequences. Further provided herein are methods wherein the method further comprises amplifying the nucleic acid. Further provided herein are methods wherein the method further comprises amplifying a single strand of the nucleic acid. Further provided herein are methods wherein at least 50, 75, 90, 95, 98, or at least 99% of the amplicons have a length corresponding to the total length of all fragment regions and stabilizing regions in the composition. Further provided herein are methods wherein at least 50, 75, 90, 95, 98, or at least 99% of the amplicons have a length corresponding to the total length of all fragment regions and stabilizing regions in the composition without purification. Further provided herein are methods wherein at least 50, 75, 90, 95, 98, or at least 99% of the amplicons have a length corresponding to the total length of all fragment regions and stabilizing regions in the composition without digestion, filtration, size purification, or gel purification. Further provided herein are methods wherein the method further comprises purifying the nucleic acid. Further provided herein are methods wherein the polynucleotide fragments are synthesized. Further provided herein are methods wherein the polynucleotide fragments are synthesized on a surface. Further provided herein are methods wherein the polynucleotide fragments are synthesized using chemical synthesis or template-free enzymatic synthesis. Further provided herein are methods wherein assembly occurs at 40-95 degrees Celsius. Further provided herein are methods wherein assembly has a complexity score of at least 5, 10, 15, 20, 30, 50, or at least 100. Further provided herein are methods wherein assembly has a complexity score of at least 5, 10, 15, 20, 30, 50, or at least 100 when measured using an in-WSGR Docket No.: 69686-701.601silico assembly tool. Further provided herein are methods wherein the two or more polynucleotide fragments comprise a composition disclosed herein. Further provided herein are methods wherein the nucleic acid is assembled in no more than 1 day, 1 week, 2 weeks, or 4 weeks. Further provided herein are methods wherein the nucleic acid is assembled at a rate of at least 30kb, lOOkb, or 1Mb per week. Further provided herein are methods wherein the method further comprises inserting the nucleic acid into a cell, genome, artificial chromosome, plasmid, or vector. Further provided herein are methods wherein inserting comprises recombination, TALEN, CRISPR, or Zn-finger endonucleases.

[0007] Provided herein are systems for nucleic acid assembly comprising: an input comprising a nucleic acid sequence of nucleic acid to be assembled; and one or more computer-implemented algorithms configured to: generate two or more fragments of the nucleic acid sequence; append index sequences to a portion of the fragments; and output a library of synthetic sequences, wherein a portion of the synthetic sequences comprise the index sequences. Further provided herein are systems wherein the system further comprises instructions to synthesize polynucleotide fragments corresponding to the synthetic sequences in the library. Further provided herein are systems wherein the system further comprises a device configured to synthesize the polynucleotide fragments corresponding to the synthetic sequences in the library. Further provided herein are systems wherein the system further comprises a device configured to assemble the polynucleotide fragments corresponding to the synthetic sequences in the library. Further provided herein are systems wherein the device is configured to execute a method described herein. Further provided herein are systems wherein the index sequences are less than 90, 95%, 97%, or 99% identical or complementary to any fragment region in the polynucleotide library. Further provided herein are systems wherein each index sequences is different than any contiguous sequence in the nucleic acid sequence. Further provided herein are systems wherein each index sequence has a Hamming or Levenshtein distance of at least two from any other index sequence. Further provided herein are systems wherein the number of index sequences equals the number of fragments minus two. Further provided herein are systems wherein the computer-implemented algorithms comprise use of machine learning. Further provided herein are systems wherein machine learning comprises inductive learning (e.g., decision / classification trees such as RF (random forest), C&RT, boosted trees, etc.), Probably Approximately Correct(PAC) learning, connect! onist learning (e.g., neural networks (NN), artificial neural networks (ANN), neuro fuzzy networks (NFN), network structures, the Cox Proportional-Hazards Model (CPHM), perceptrons such as multi-layer perceptrons, multi-layer feed-forwardWSGR Docket No.: 69686-701.601networks, applications of neural networks, Bayesian learning in belief networks, reinforcement learning, support vector machines (e.g., Kernel methods), multivariate adaptive regression splines (MARS), Levenberg-Marquardt algorithms, Gauss-Newton algorithms, mixtures of Gaussians, gradient descent algorithms, or learning vector quantization (LVQ). Further provided herein are systems wherein the library of synthetic sequences comprises at least 5, 10, 15, 20, 30, 40, 50, 75, 100, or at least 500 synthetic sequences.

[0008] Provided herein are synthetic biological entity comprising a synthetic genome, wherein the synthetic genome is at least 1, 5, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100 Mb or more in size; comprises no more than 10, 20, 30, 40, 50, 60, 70, or no more than 80% sequence homology with any naturally occurring genome; or comprises no more than 5, 10, 15, 20, 25, 30, 40, 50, or no more than 60% of its genes or non-coding regions derived from any single naturally occurring biological entity; and the synthetic biological entity is distinguishable from naturally occurring biological entities due to one or properties. Further provided herein are synthetic biological entities wherein the biological entity is distinguishable from naturally occurring biological entities due to at least 1, 2, 3, 5, 10, 20, or at least 30 properties. Further provided herein are synthetic biological entities wherein the genome comprises coding or non-coding DNA regions from at least 5, 10, 15, 20, 25, 30 or at least 30 different biological entities. Further provided herein are synthetic biological entities wherein the biological entity comprises a bacterium, yeast, or virus. Further provided herein are synthetic biological entities wherein the biological entity comprises a synthetic bacterium and the one or more properties are: cell morphology, colony morphology, reproduction, metabolism, growth conditions, stress resistance, pathogenicity, cell wall composition, or secondary metabolites. Further provided herein are synthetic biological entities wherein the biological entity comprises a synthetic virus and the one or more properties are: morphology, genome type, envelope presence, replication cycle, host range, transmission mode, cytopathic effects, immune evasion, or pathogenicity. Further provided herein are synthetic biological entities wherein the biological entity comprises a synthetic yeast and the one or more properties are: cell morphology; colony morphology reproduction, metabolism, growth conditions, stress resistance, pathogenicity, cell wall composition, or secondary metabolites.

[0009] Provided herein are methods of synthesizing a synthetic biological entity provided herein comprising synthesizing the synthetic genome and incorporating the synthetic genome or iteratively incorporating portions thereof into a cell or other vector.WSGR Docket No.: 69686-701.601INCORPORATION BY REFERENCEAll publications, patents, and patent applications mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent, or patent application was specifically and individually indicated to be incorporated by reference.BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 illustrates a workflow for constructing large DNA and the design and construction of de novo DNA, up to and including genome scale.

[0011] Figure 2 depicts some uses for the methods and systems described herein.

[0012] Figures 3A-3B depicts methods of nucleic acid assembly. Figure 3A depicts methods of nucleic acid assembly with PC A, Gibson Assembly, Golden Gate, USER, Homologous Recombination and Yeast Assembly. Figure 3B depicts misassembly errors that arise using methods of nucleic acid assembly other than the current method.

[0013] Figure 4A depicts a standard method of nucleic acid assembly, such as PCA.

[0014] Figure 4B depicts a method of nucleic acid assembly described herein.

[0015] Figure 5A depicts a method of nucleic acid assembly using 3 -way junctions. The stabilizing region is indicated as a and a*. The index region is indicated as b and b*.

[0016] Figure 5B depicts the scalability of the method of nucleic acid assembly of Figure 5A.DETAILED DESCRIPTION

[0017] In some instances, the methods and systems described herein relate to the interaction between Artificial intelligence (Al)-informed design and corresponding DNA construction. For instance, FIG. 1 depicts a workflow for using Al-informed DNA design in de novo genome construction. First, Al-informed design is used for gene construction. In combination with the deep sequence space provided by the methods and systems described herein, a generative model for DNA design is created. This then creates Artificial Biological Intelligence “ABI” which allows for genome rewriting and de novo genome rewriting. In some instances ABI comprises the ability to predict the species or function that a particular DNA sequences computes generate DNA sequences that compute a predefined biological outcome, the ability to construct DNA sequences as scale, and the ability to boot up such sequences FIG. 2 depicts some nonlimiting examples of the uses of the systems and methods described herein.

[0018] In some instance, the systems and methods can be used to deconvolute GWAS hits. In some instances, deconvoluting GWAS hits comprises the process of identifying which regionWSGR Docket No.: 69686-701.601of sequence within a GWAS hit is of functional consequence. A Genome Wide Association Study (GWAS) may detect associations between genetic variants and traits in a population sample. Characterization of a GWAS hit may enable better understanding of the biology of a disease mechanism, and identify a suitable target for applicable treatments. A GWAS can utilize genotyping and / or sequencing data, and often involves an evaluation of millions of genetic variants that are relatively evenly distributed across the genome. For instance, the methods and systems of sequence assembly and library synthesis described herein can be used to design, synthesize and test all the distributed genetic variants in order to identify which regions within the sequence of the identified GWAS hits represent causal variants that are of potential therapeutic significance.

[0019] In some embodiments, the systems and methods described herein are used in the AI=informed design of proteins or genomes. For instance, Al foundational models, such as those using deep learning and neural networks, may be used to design libraries of potential proteins, which are then constructed using the methods described herein. In some instances, Al is used to design libraries of phage genomes, both extant or extinct. Bacteriophages or “phages” represent a group of bacterial viruses and are engineered or sourced from environmental sources. Bacteriophages are bacterial viruses that rely on the host's cellular machinery to replicate. In some embodiments, the bacteriophages disclosed herein treat or prevent diseases or conditions mediated or caused by bacteria as disclosed herein in a human or animal subject. In some embodiments, the bacteriophages disclosed herein treat or prevent diseases or conditions caused or exacerbated by bacteria as disclosed herein in a human or animal subject. In some embodiments, the phage are used to target antibacterial -resistant bacteria. In some embodiments, the antibacterial-resistant bacteria are resistant to one, two, three, or more antimicrobial agents. In some embodiments, the antibacterial-resistant bacteria are resistant to all known antimicrobial agents.

[0020] In some instances, the methods and systems described herein can be used to design and optimize gene circuits. In some embodiments, the gene circuits comprise regulatory proteins, such as transcription factors. In some embodiments, the gene circuits comprise long noncoding MicroRNAs (miRNAs), Small Interfering RNAs (siRNAs), Circular RNAs (crRNAs), Piwi-Interacting RNAs (piRNAs), enhancer RNAs (eRNAs), Transfer RNA-derived Fragments (tRFs) and tRNA-derived Stress-Induced RNAs (tiRNAs), Riboswitches, Small Nucleolar RNAs (snoRNAs), Small Nuclear RNAs (snRNAs), or combinations thereof, or with protein transcription factors. In some embodiments, the gene circuits comprise long noncoding RNA (IncRNA). In some embodiments, the gene circuits compriseWSGR Docket No.: 69686-701.601long noncoding RNA (IncRNA) in combination with any of the above protein or RNA regulatory factors. In some embodiments, the methods and systems described herein may be used to design and synthesize circuits organized into a regulatory structure that embodies Boolean logic to derive specific output responses in response to certain inputs.

[0021] In some embodiments, the methods and systems described herein may be used to design and synthesize libraries in order to develop de novo gene regulatory circuits. The libraries may comprise IncRNA, protein regulatory elements, RNA regulatory elements, or combinations thereof. Regulatory circuits in some instances comprise promoters, mRNA binding regions, coding regions, terminators, or other element. In some instances, regulatory circuits comprise feedback mechanisms to control transcription or translation. In some instances, gene regulatory circuits comprise cellular circuits (both intracellular and extracellular circuits). For instance, a cellular, e.g., regulatory circuit combines trans inputs (such as the levels and activities of factors, e.g., transcription factors, non-coding RNAs, e.g., regulatory RNAs and signaling molecules) and cis inputs (such as sequences, e.g., regulatory sequences in the promoter and enhancer of a gene).

[0022] In some instances, the methods and systems described herein can be used to design and optimize gene therapy. There are different methods of gene therapy, including gene replacement, gene editing, base editing, epigenetic editing, and RNA editing. As used herein, “gene replacement” comprises a system that delivers a therapeutic gene to a subject. The therapeutic gene may comprise a gene which is overexpressed. The therapeutic gene may suppress a relevant target or phenotype. The therapeutic gene may be integrated into the genome. The therapeutic gene may remain in the cell as an episome. As used herein, “gene editing” comprises altering a sequence of the target gene within the genome. Gene editing may refer to replacing a sequence with a different sequence. Gene editing may refer to inducing mutations in a genomic sequence. Gene editing may refer to producing a knock-out version of a genomic sequence. It may employ one of several different technologies including CRISPR, ZFPs, or transposons delivered using AAV, lipid nanoparticles or other methods. As used herein, “RNA editing” comprises altering a sequence of an RNA within in an organism.I. Methods for Sequence Assembly

[0023] Described herein are methods and compositions for the assembly of nucleic acid sequences. Assembly of such sequences may in some cases be challenging due to specific properties of the assembly fragments, such as GC content, repeating regions, and secondary structure. Additionally, assembly of libraries of such sequences may be assembled in parallel,WSGR Docket No.: 69686-701.601with members of the library possessing regions of high variability across members. Such parallel assembly of fragments is challenging due to the presence of highly variable regions across members of the library for such fragments. Moreover, assembly may result in errors, such as incorrectly assembled nucleic acids. Nucleic acids comprising variable regions may include nucleic acids encoding for genes (such as proteins or antibodies), or non-coding nucleic acids. In some instances, a nucleic acid assembled herein comprises a region encoding for an immunoglobulin or fragment thereof. Assembly of libraries comprising nucleic acids of high variability may be accomplished by the methods described herein.

[0024] FIG. 3A depicts methods of nucleic acid assembly including PC A. Gibson Assembly, Golden Gate, USER, Homologous Assembly and Yeast assembly. In these methods, a double stranded fragment X and a double stranded fragment Y are combined with the use of adapters a and a’ . the top strand of fragment X is ligated to adapter a and the bottom strand of fragment Y is ligated to adapter a’. Adapters a and a’ may comprise homologous sequences and anneal to form a double stranded sequence with two nicks. The double stranded sequence is then ligated to produce a fragment comprising the sequences of fragment X, the adapters and fragment Y. These methods can result in some degree of misassembly, as depicted in FIG. 3B. For instance, the adapter may form a secondary structure and result in decreased target product yield. Fragments may also misassociate or misalign if adapters incorrectly anneal with each other. These incorrect interactions increase exponentially with the number of fragments used in assembly. These methods also may struggle with assembling nucleic acids with complex structures. Additionally, since the adaptors are incorporated into the final sequence, there are limitations in their design. FIG. 4A is another depiction of a method of nucleic acid assembly. A target synthetic sequence is bioinformatically split and the individual oligos are synthesized then annealed.

[0025] In contrast, the methods described herein may include those depicted in FIGS. 5A-5B. In one method, fragment X comprises a double stranded polynucleotide with a first strand and a second strand. The first strand comprises a stabilizing region (a) and an index region (b). In some instances, the index region comprises a nucleic acid, protein, or small molecule. Fragment Y comprises a double stranded polynucleotide with a first strand and a second strand. The first strand comprises a second index region b* and third strand comprises a second stabilizing region b*. The two index regions anneal and the two stabilizing regions anneal to form a 3 -way junction. In some instances, nucleic acids, proteins, or small molecule of one index region of a first fragment binds with the nucleic acids, proteins, or small molecule of an index region of another fragment. The strands are ligated to connect theWSGR Docket No.: 69686-701.601assembly fragments. At least in part because the index regions are not incorporated into the final fragment, they can be unique and increase in length to increase fidelity. This assembly process scales with improved fidelity. The nucleic acid sequence for synthesis may be bioinformatically split as in FIG. 4B. The fragment specific barcodes or index regions may be designed using artificial intelligence. In some instances, the methods include those described in Robinson, N.E., Zhang, W., Ghosh, R. etal. Construction of complex and diverse DNA sequences using DNA three-way junctions. Nature 651, 491-500 (2026), which is herein incorporated by reference, including the Sidewinder 3-way junction assembly method.

[0026] Provided herein are compositions wherein the index region comprises a protein. In some instances, the protein is configured to bind to a small molecule (e.g., streptavidin / biotin), or to a nucleic acid (e.g., a transcription factor and a promoter).

[0027] Provided herein are compositions wherein the index region comprises a nucleic acid. In some embodiments, index sequences comprise canonical nucleic acids. In some instances, index sequences comprise non-canonical nucleic acids (see US20190376054, incorporated by reference in its entirety). In some instances, the methods shown in FIG. 3A are used alone or in combination with assembly methods described herein. In some instances, assembly methods comprise PCA, Gibson Assembly, Golden Gate, USER, Homologous Recombination and Yeast Assembly. In some instances, methods comprise barcode-based assembly such as those described in US20210071168, incorporated by reference in its entirety. In some instances, methods comprise

[0028] In some aspects, the methods comprise assembling two or more polynucleotide fragments into the nucleic acid, wherein the nucleic comprises one or more of complex regions. The one or more complex regions individually or collectively comprise secondary structure (optionally alpha helix, beta sheet, stem-loop, pseudoknot, homodimer, heterodimer, hairpin loop, interior loop, bulge, or multiloop), inversions, palindromic sequences, repeats, high GC, or low AT. In some embodiments, the nucleic acid comprises at least two complex regions. In some embodiments, the nucleic acid comprises at least 1, 2, 5, 10, 20, 30, or 40% complex regions.

[0029] In some embodiments, the method assembles at least 5, 10, 15, 20, 25, 30, 40, 50 or at least 100 polynucleotide fragments. In some embodiments, the nucleic acid is at least 1000, 3000, 5000, 10,000, 20,000, 50,000, 100,000, 500,000, 1 million, 10 million, 100 million, 1 billion, or at least 10 billion nucleotides in length. In some embodiments, the nucleic acid isWSGR Docket No.: 69686-701.6011000-5000, 2000-5000, 2000-10,000, 5000-10,000, 10,000-100,000, 10,000-75,000, 25,000-100,000, 25,000-75,000 or 10,000 to 50,000 nucleotides in length.

[0030] In some embodiments, the assembling comprises contacting two or more polynucleotide fragments with a nick repair enzyme. In some embodiments, the assembling comprises contacting two or more polynucleotide fragments with a ligase. The methods may further comprise amplifying the nucleic acid. In some embodiments, a single strand of the nucleic acid is amplified.

[0031] In some embodiments, at least 50, 75, 90, 95, 98, or at least 99% of the amplicons have a length corresponding to the total length of all fragment regions and stabilizing regions in the composition. In some embodiments, at least 50, 75, 90, 95, 98, or at least 99% of the amplicons have a length corresponding to the total length of all fragment regions and stabilizing regions in the composition without purification. In some embodiments, at least 50, 75, 90, 95, 98, or at least 99% of the amplicons have a length corresponding to the total length of all fragment regions and stabilizing regions in the composition without digestion, filtration, size purification, or gel purification. In some embodiments, the length of each fragment collectively encoding for a nucleic acid to be assembled is 50-300, 50-200, 50-150, 100-500, 100-300, 100-200, 200-500, 200-300, 300-500, or 500-1000 bases. In some embodiments, a library comprises fragments having no more than 1, 2, 3, 5, 10, 15, or no more than 20% variation in length. In some embodiments, the method further comprises purifying the nucleic acid. In some embodiments, the polynucleotide fragments are synthesized using the systems and methods described herein. In some embodiments, the polynucleotide fragments are synthesized on a surface. In some embodiments, the polynucleotide fragments are synthesized using chemical synthesis or template-free enzymatic synthesis.

[0032] The methods and systems described herein include assembly of nucleic acid sequences. In some embodiments, assembly occurs at 40-95 degrees Celsius. In some embodiments, assembly occurs at 40-90, 40-80, 40-70, 40-60, 50-90, 50-80, 50-70, 50-60, 55-75, 60-65, 60-70, 60-80, 70-80, 70-90, 80-95, or 40-95 degrees Celsius. In some instances, assembly occurs under essentially isothermal conditions.

[0033] The methods and system described herein can be used for assembly of complex sequences. In some instances, assembly has a complexity score of at least 5, 10, 15, 20, 30, 50, or at least 100. In some instances, assembly has a complexity score of at least 5, 10, 15, 20, 30, 50, or at least 100 when measured using an in-silico assembly tool. In someWSGR Docket No.: 69686-701.601embodiments, the complexity tool comprises the synthesis input tool provided by Integrated DNA Technologies.

[0034] The methods and systems described herein may have advantages over methods of assembly, including an improvement in the rate of assembly. In some embodiments, the nucleic acid is assembled in no more than 1 day, 1 week, 2 weeks, or 4 weeks. In some embodiments, the nucleic acid is assembled at a rate of at least 30kb, lOOkb, or 1Mb per week.

[0035] In some embodiments, the method further comprises inserting the nucleic acid into a cell, genome, artificial chromosome, plasmid, or vector. In some embodiments, inserting comprises recombination, TALEN, CRISPR, or Zn-finger endonucleases. In some embodiments, the nucleic acid originates from an animal, plant, bacteria, virus, fungi, or archaea. In some embodiments, the nucleic acid originates from a eukaryote. In some embodiments, the nucleic acid encodes a protein, a regulatory element, digital information, genome, or vector.

[0036] In some embodiments, the nucleic acid encodes for a protein, wherein the protein optionally comprises a collagen, elastin, rubisco, cellulase, a fibroblast growth factor (FGF), an antibody, or telomerase. The protein may comprise collagen. The collagen may be COLA1 A. The protein may comprise elastin. The protein may comprise rubisco. The protein may comprise cellulase. The protein may comprise an FGF. The FGF may comprise FGF2. The protein may comprise a telomerase. In some embodiments, the nucleic acid includes highly repetitive sequences. Nonlimiting examples of nucleic acids with highly repetitive sequences include collagens, elastin, and rubisco. In some embodiments, the nucleic acid may include GC rich regions. Nonlimiting examples of nucleic acids with GC rich regions include cellulase, FGFs, and telomerase. In some embodiments, the nucleic acid encodes for a genome, wherein the genome comprises at least 10, 20, 30, 40, 50, 70, or at least 80% repetitive sequences.

[0037] In some embodiments, the composition collectively encodes for an antibody library having a diversity of at least 106, 107, 108, 109, 1010, 1011or at least 1012unique members. In some instances the antibody library encodes for antibodies or fragments there. In some instances an antibody fragment may be a nanobody, Fab, Fab', (Fab')2, Fv, ScFv, diabody, triabody, tetrabody, Bis-scFv, minibody, Fab2, or Fab3 fragment. Libraries in some instances comprise a variant nucleic acid library. Variant TCR sequence libraries described herein may comprise common sequence from any region of a TCR nucleotide sequence encoding for a gene or gene fragment. Exemplary regions of TCR nucleotide sequence for gene or geneWSGR Docket No.: 69686-701.601fragment variation include, without limitation: variable region of the a chain, variable region of the P chain, constant region of the a chain, constant region of the P chain, hinge region, transmembrane region, or cytoplasmic tail region. In some instances, thevariant TCR sequence library comprises common sequence for variation, wherein nucleic acids encoding for residues outside of the complementarity determining region of the variable domains are selected for variation. In some instances, the variant TCR sequence library comprises common sequence for variation, wherein nucleic acids encoding for residues within the complementarity determining region of the variable domains are selected for variation. Sequence encoding for the TCR a and beta chains each comprise separate variable (V), diversity (D), joining (J) gene segments, and constant (C) genes. The TCR a locus (chromosome 14) consists of 70-80 Va gene segments, each preceded by an exon encoding the leader sequence (L). A cluster of 61 Ja gene segments is located a considerable distance from the Va gene segments. The Ja gene segments are followed by a single C gene, which contains separate exons for the constant and hinge domains and a single exon encoding the transmembrane and cytoplasmic regions (not shown). The TCRP locus (chromosome 7) has a different organization, with a cluster of 52 functional VP gene segments located distantly from two separate clusters each containing a single D gene segment, together with six or seven J gene segments and a single C gene. Each TCRP C gene has separate exons encoding the constant domain, the hinge, the transmembrane region, and the cytoplasmic region (not shown). The TCRa locus is interrupted between the J and V gene segments by another T-cell receptor locus — the TCR.6 locus. In some embodiments, the common sequence selected for variation includes one or more of the Va gene segment, Ja gene segment, Ca gene segment, VP gene segment, jp gene segment or CP gene segment.

[0038] In some instances, a number of final constructs is generated. In some instances, final constructs are ligated together to form a genome using the methods describe herein. In some instances, a library comprises one or more final constructs. In some instances, a number of final constructs generated is 1000-10000, 1000-100,000, 1000-50,000, 1000-25,000, 1000-10,000, or 1000 to 5000. In some instances, a number of final constructs generated is about 1000, 2000, 5000, 8000, 10000, 15,000, 20,000, 100,000, or about 1,000,000. In some instances, a number of final constructs generated is at least 1000, 2000, 5000, 8000, 10000, 15,000, 20,000, 100,000, or at least 1,000,000.

[0039] Described herein are methods of de novo synthesis for nucleic acid sequence assembly. Such methods are in some instances used for the assembly of smaller nucleic acid fragments. In some instances, nucleic acid fragments comprise constant regions, variableWSGR Docket No.: 69686-701.601regions, overlap regions, hypervariable regions, barcodes, regions encoding for peptide cleavage sites, regions encoding for genes or fragments of genes, restriction sites, or other region.

[0040] Methods described herein for nucleic acid assembly may comprise a ligation reaction. In some instances, methods described herein comprise an amplification reaction. In some instances, the amplification reaction is polymerase chain reaction (PCR). In some instances, the amplification reaction is dial-out PCR. In some instances, the amplification reaction comprises hybridization of a universal primer binding sequence during amplification. In some instances, the universal primer binding sequence is capable of binding the same 5’ or 3’ primer. In some instances, the universal primer binding sequence is shared among a plurality of target nucleic acids in the amplification reaction.

[0041] The resulting nucleic acids can be verified. In some cases, the nucleic acids are verified by sequencing. In some instances, the nucleic acids are verified by high-throughput sequencing such as by next generation sequencing. Sequencing of the sequencing library can be performed with any appropriate sequencing technology, including but not limited to single-molecule real-time (SMRT) sequencing, Polony sequencing, sequencing by ligation, reversible terminator sequencing, proton detection sequencing, ion semiconductor sequencing, nanopore sequencing, electronic sequencing, pyrosequencing, Maxam-Gilbert sequencing, chain termination (e.g., Sanger) sequencing, +S sequencing, or sequencing by synthesis.

[0042] Methods as described herein, in some embodiments, result in generation of libraries comprising at least or about 101, 102, 103, 104, 105, 106, 107, 108, 109, 1010, or more than 1010variants. In some instances, sequences for each variant of the libraries comprising at least or about 101, 102, 103, 104, 105, 106, 107, 108, 109, or 1010variants are known. In some instances, the libraries comprise a predicted diversity of variants. In some instances, the diversity represented in the libraries is at least or about 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, or more than 95% of the predicted diversity. In some instances, the diversity represented in the libraries is at least or about 70% of the predicted diversity. In some instances, the diversity represented in the libraries is at least or about 80% of the predicted diversity. In some instances, the diversity represented in the libraries is at least or about 90% of the predicted diversity. In some instances, the diversity represented in the libraries is at least or about 99% of the predicted diversity. As described herein the term “predicted diversity” refers to a total theoretical diversity in a population comprising all possible variants.WSGR Docket No.: 69686-701.601

[0043] Nucleic acid assembly using methods as described herein may efficiently assemble fragments despite high GC content, direct repeats, or secondary structures. In some instances, the fragments for assembly comprise GC content of at least or about 5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, or more than 95%. In some instances, the fragments for assembly comprise at least or about 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, or 80 base pairs (bp) adjacent direct repeats. In some instances, the fragments for assembly comprise secondary structures such as hairpin structures with dG values of at least or about -5, -6, -7, -8, -9, -10, -11, -12, -13, -14, -15, -16, -17, -18, -19, -20, -21, -22, -23, -24, -25, or -26 dG. In some instances, the fragments for assembly comprise secondary structures such as hairpin structures with dG values in a range of about -11 to about -18 dG.II. Systems for Nucleic Acid Sequence Assembly

[0044] In certain aspects, disclosed herein is a system for nucleic acid assembly comprising an input and one or more computer-implemented algorithms. In some embodiments, the input comprises a nucleic acid sequence of nucleic acid to be assembled. In some embodiments, the one or more computer-implemented algorithms are configured to generate two or more fragments of the nucleic acid sequence; append index sequences to a portion of the fragments; and output a library of synthetic sequences, wherein a portion of the synthetic sequences comprise the index sequences.

[0045] In some embodiments, the system further comprises instructions to synthesize polynucleotide fragments corresponding to the synthetic sequences in the library. In some embodiments, the system further comprises a device configured to synthesize the polynucleotide fragments corresponding to the synthetic sequences in the library. In some embodiments, the system further comprises a device configured to assemble the polynucleotide fragments corresponding to the synthetic sequences in the library. In some embodiments, the device is configured to execute the method described herein.

[0046] In some embodiments, the index sequences are less than 90, 95%, 97%, or 99% identical or complementary to any fragment region in the polynucleotide library. In some embodiments, each index sequences is different than any contiguous sequence in the nucleic acid sequence. In some embodiments, each index sequence has a Hamming or Levenshtein distance of at least two from any other index sequence. In some embodiments, the number of index sequences equals the number of fragments minus two.WSGR Docket No.: 69686-701.601A. Polynucleotide libraries

[0047] In certain aspects, described herein are polynucleotide libraries. In some embodiments, the polynucleotide libraries comprise at least a first and a second polynucleotide. In some embodiments, the first polynucleotide comprises a first strand and a second strand, the first strand comprises a first fragment region, a first stabilizing region, and a first index region; the second strand comprises a second fragment region, wherein the first fragment region and the second fragment region are at least partially complementary. In some embodiments, the second polynucleotide, comprises a third strand and a fourth strand, wherein the third strand comprises a second index region and a third fragment region, and the fourth strand comprises a second stabilizing region and a fourth fragment region; wherein the third fragment region and the fourth fragment region are at least partially complementary. In some embodiments, the first index region and the second index region are at least partially complementary, and the first stabilizing region and the second stabilizing region are at least partially complementary.

[0048] In some embodiments, one or more of the first fragment region, the second fragment region, the third fragment region, and the fourth fragment region is independently 100-1000, 100-2500, 500-5000, 500-10,000, or 500-100,000 nucleotides in length.

[0049] In some embodiments, the index sequences is independently 10-200, 10-150, 10-100, 10-50, or 10-25 nucleotides in length. In some embodiments, one or more of the stabilizing sequences is independently 10-200, 10-150, 10-100, 10-50, or 10-25 nucleotides in length.

[0050] In some embodiments, the first stabilizing region and the second stabilizing region have a Tmof about 40 °C to about 95 °C. In some embodiments, the first stabilizing region and the second stabilizing region have a Tmof about 40 °C to about 45 °C, about 40 °C to about 50 °C, about 40 °C to about 55 °C, about 40 °C to about 60 °C, about 40 °C to about 65 °C, about 40 °C to about 70 °C, about 40 °C to about 75 °C, about 40 °C to about 80 °C, about 40 °C to about 85 °C, about 40 °C to about 90 °C, about 40 °C to about 95 °C, about 45 °C to about 50 °C, about 45 °C to about 55 °C, about 45 °C to about 60 °C, about 45 °C to about 65 °C, about 45 °C to about 70 °C, about 45 °C to about 75 °C, about 45 °C to about 80 °C, about 45 °C to about 85 °C, about 45 °C to about 90 °C, about 45 °C to about 95 °C, about 50 °C to about 55 °C, about 50 °C to about 60 °C, about 50 °C to about 65 °C, about 50 °C to about 70 °C, about 50 °C to about 75 °C, about 50 °C to about 80 °C, about 50 °C to about 85 °C, about 50 °C to about 90 °C, about 50 °C to about 95 °C, about 55 °C to about 60 °C, about 55 °C to about 65 °C, about 55 °C to about 70 °C, about 55 °C to about 75 °C, about 55 °C to about 80 °C, about 55 °C to about 85 °C, about 55 °C to about 90 °C, about 55 °C to about 95 °C, aboutWSGR Docket No.: 69686-701.60160 °C to about 65 °C, about 60 °C to about 70 °C, about 60 °C to about 75 °C, about 60 °C to about 80 °C, about 60 °C to about 85 °C, about 60 °C to about 90 °C, about 60 °C to about 95 °C, about 65 °C to about 70 °C, about 65 °C to about 75 °C, about 65 °C to about 80 °C, about 65 °C to about 85 °C, about 65 °C to about 90 °C, about 65 °C to about 95 °C, about 70 °C to about 75 °C, about 70 °C to about 80 °C, about 70 °C to about 85 °C, about 70 °C to about 90 °C, about 70 °C to about 95 °C, about 75 °C to about 80 °C, about 75 °C to about 85 °C, about 75 °C to about 90 °C, about 75 °C to about 95 °C, about 80 °C to about 85 °C, about 80 °C to about 90 °C, about 80 °C to about 95 °C, about 85 °C to about 90 °C, about 85 °C to about 95 °C, or about 90 °C to about 95 °C. In some embodiments, the first stabilizing region and the second stabilizing region have a Tmof about 40 °C, about 45 °C, about 50 °C, about 55 °C, about 60 °C, about 65 °C, about 70 °C, about 75 °C, about 80 °C, about 85 °C, about 90 °C, or about 95 °C. In some embodiments, the first stabilizing region and the second stabilizing region have a Tmof at least about 40 °C, about 45 °C, about 50 °C, about 55 °C, about 60 °C, about 65 °C, about 70 °C, about 75 °C, about 80 °C, about 85 °C, or about 90 °C. In some embodiments, the first stabilizing region and the second stabilizing region have a Tmof at most about 45 °C, about 50 °C, about 55 °C, about 60 °C, about 65 °C, about 70 °C, about 75 °C, about 80 °C, about 85 °C, about 90 °C, or about 95 °C.

[0051] In some embodiments, the first index region and the second index region have a Tmof about 40 °C to about 95 °C. In some embodiments, the first index region and the second index region have a Tmof about 40 °C to about 45 °C, about 40 °C to about 50 °C, about 40 °C to about 55 °C, about 40 °C to about 60 °C, about 40 °C to about 65 °C, about 40 °C to about 70 °C, about 40 °C to about 75 °C, about 40 °C to about 80 °C, about 40 °C to about 85 °C, about 40 °C to about 90 °C, about 40 °C to about 95 °C, about 45 °C to about 50 °C, about 45 °C to about 55 °C, about 45 °C to about 60 °C, about 45 °C to about 65 °C, about 45 °C to about 70 °C, about 45 °C to about 75 °C, about 45 °C to about 80 °C, about 45 °C to about 85 °C, about 45 °C to about 90 °C, about 45 °C to about 95 °C, about 50 °C to about 55 °C, about 50 °C to about 60 °C, about 50 °C to about 65 °C, about 50 °C to about 70 °C, about 50 °C to about 75 °C, about 50 °C to about 80 °C, about 50 °C to about 85 °C, about 50 °C to about 90 °C, about 50 °C to about 95 °C, about 55 °C to about 60 °C, about 55 °C to about 65 °C, about 55 °C to about 70 °C, about 55 °C to about 75 °C, about 55 °C to about 80 °C, about 55 °C to about 85 °C, about 55 °C to about 90 °C, about 55 °C to about 95 °C, about 60 °C to about 65 °C, about 60 °C to about 70 °C, about 60 °C to about 75 °C, about 60 °C to about 80 °C, about 60 °C to about 85 °C, about 60 °C to about 90 °C, about 60 °C to about 95 °C, about 65 °C to about 70 °C, about 65 °C to about 75 °C, about 65 °C to about 80 °C, about 65 °C to about 85 °C, aboutWSGR Docket No.: 69686-701.60165 °C to about 90 °C, about 65 °C to about 95 °C, about 70 °C to about 75 °C, about 70 °C to about 80 °C, about 70 °C to about 85 °C, about 70 °C to about 90 °C, about 70 °C to about 95 °C, about 75 °C to about 80 °C, about 75 °C to about 85 °C, about 75 °C to about 90 °C, about 75 °C to about 95 °C, about 80 °C to about 85 °C, about 80 °C to about 90 °C, about 80 °C to about 95 °C, about 85 °C to about 90 °C, about 85 °C to about 95 °C, or about 90 °C to about 95 °C. In some embodiments, the first index region and the second index region have a Tmof about 40 °C, about 45 °C, about 50 °C, about 55 °C, about 60 °C, about 65 °C, about 70 °C, about 75 °C, about 80 °C, about 85 °C, about 90 °C, or about 95 °C. In some embodiments, the first index region and the second index region have a Tmof at least about 40 °C, about 45 °C, about 50 °C, about 55 °C, about 60 °C, about 65 °C, about 70 °C, about 75 °C, about 80 °C, about 85 °C, or about 90 °C. In some embodiments, the first index region and the second index region have a Tmof at most about 45 °C, about 50 °C, about 55 °C, about 60 °C, about 65 °C, about 70 °C, about 75 °C, about 80 °C, about 85 °C, about 90 °C, or about 95 °C.

[0052] In some embodiments, the first index region and the second index region are less than 90, 95%, 97%, or 99% identical or complementary to any fragment region in the polynucleotide library. In some embodiments, the first index region and the second index region are different than any fragment region in the polynucleotide library. In some embodiments, any index region in the library comprises a Hamming or Levenshtein distance of at least two from any other index region.

[0053] In some embodiments, the second stabilizing region and the second fragment region are covalently attached. In some embodiments, the first strand comprises from 5’ to 3’: the first fragment region, the first stabilizing region, and the first index region. In some embodiments, the first strand further comprises a 5’ third index sequence. In some embodiments, the first strand comprises a 3’ overhang. In some embodiments, the 3’ overhang comprises the first stabilizing region and the first index region. In some embodiments, the second strand further comprises a third 5’ stabilizing sequence. In some embodiments, the second nucleic acid comprises at least one non- complementary region. In some embodiments, the at least one non-complementary region is formed between the second index sequence and the second stabilizing region. In some embodiments, the third strand comprises from 5’ to 3’: the second index region and the third fragment region. In some embodiments, the third strand comprises a 5’ overhang. In some embodiments, the 5’ overhang comprises the second index region. In some embodiments, the fourth strand comprises from 5’ to 3’: the fourth fragment region and the second stabilizing region. In some embodiments, one or more of the fragment regions comprise one or more of complexWSGR Docket No.: 69686-701.601regions. In some embodiments, the one or more of the fragment regions individually or collectively comprise at least two complex regions. In some embodiments, the one or more of the fragment regions individually or collectively comprise at least 1, 2, 5, 10, 20, 30, 40, 50, 60, 75, 80, or at least 90% complex regions. In some embodiments, the one or more complex regions individually or collectively comprise secondary structure (optionally alpha helix, beta sheet, stem-loop, pseudoknot, homodimer, heterodimer, hairpin loop, interior loop, bulge, or multiloop), inversions, palindromic sequences, repeats, high GC, or low AT. In some embodiments, the composition does not comprise one or more of a polymerase, exonuclease, endonuclease, SSB, or recombinase. In some embodiments, the 5’ end of the first fragment region and the 3’ end of the third fragment region are not substantially complementary.

[0054] In some embodiments, the library comprises at least 5, 10, 15, 20, 30, 40, 50, 75, 100, or at least 500 unique polynucleotides.B. Vectors and Cells

[0055] The methods disclosed herein may comprise inserting the nucleic acid into a cell, genome, artificial chromosome, plasmid, or vector. A nucleic acid is a type of polynucleotide comprising two or more nucleotide bases. In certain embodiments, the nucleic acid is a component of a vector that can be used to transfer the polypeptide encoding polynucleotide into a cell. As used herein, the term “vector” refers to a nucleic acid molecule capable of transporting another nucleic acid to which it has been linked. One type of vector is a genomic integrated vector, or “integrated vector,” which can become integrated into the chromosomal DNA of the host cell. Another type of vector is an “episomal” vector, e.g., a nucleic acid capable of extra-chromosomal replication. Vectors capable of directing the expression of genes to which they are operatively linked are referred to herein as “expression vectors.” Suitable vectors comprise plasmids, bacterial artificial chromosomes, yeast artificial chromosomes, viral vectors and the like. In the expression vectors regulatory elements such as promoters, enhancers, polyadenylation signals for use in controlling transcription can be derived from mammalian, microbial, viral or insect genes. The ability to replicate in a host, usually conferred by an origin of replication, and a selection gene to facilitate recognition of transformants may additionally be incorporated. Vectors derived from viruses, such as lentiviruses, retroviruses, adenoviruses, adeno-associated viruses, and the like, may be employed. Plasmid vectors can be linearized for integration into a chromosomal location. Vectors can comprise sequences that direct site-specific integration into a defined location or restricted set of sites in the genome (e.g., AttP-AttB recombination). Additionally, vectorsWSGR Docket No.: 69686-701.601can comprise sequences derived from transposable elements. Vectors in some instances include but are not limited to Plasmid vectors (insert size of 5-25 kb), Cosmids (30-50 kb). PAC (Pl artificial chromosome, 80-100 kb,) BAC (Bacterial Artificial Chromosome, 150— 300 kb, YAC (Yeast Artificial Chromosome, 500 kb of DNA, HACs (human artificial chromosome, >2,000 kb). Vectors in some instances comprise non-viral vectors. In some instances, non-viral vectors include but are not limited to liposomes and lipid nanoparticles (LNPs), highly branched poly(P-amino ester) (HPAE), single-chain cyclic polymer (SCKP), poly (amidoamine) (PAMAM) dendrimers, and polyethyleneimine (PEI).

[0056] In some inserting embodiments, inserting comprises recombination, clustered regularly interspaced short palindromic repeats (CRISPR), Zinc Finger Nucleases (ZFNs) (see, Umov et al., Nat. Rev. Genet., 11 :636-646 (2010) or transcription activator-like effector nucleases (TALENs) (see, Joung and Sander, Nat. Rev. Mol. Cell Biol, 1 :49-55 (2013).C. Machine Learning

[0057] In some embodiments, the systems and methods described herein comprise machine learning.

[0058] In some embodiments, machine learning comprises inductive learning (e.g., decision / classification trees such as RF (random forest), C&RT, boosted trees, etc.), Probably Approximately Correct (PAC) learning, connectionist learning (e.g., neural networks (NN), artificial neural networks (ANN), neuro fuzzy networks (NFN), network structures, the Cox Proportional-Hazards Model (CPHM), perceptrons such as multi-layer perceptrons, multilayer feed-forward networks, applications of neural networks, Bayesian learning in belief networks, reinforcement learning, support vector machines (e.g., Kernel methods), multivariate adaptive regression splines (MARS), Levenberg-Marquardt algorithms, Gauss-Newton algorithms, mixtures of Gaussians, gradient descent algorithms, or learning vector quantization (LVQ).

[0059] Training the ML model may include, in some cases, selecting one or more untrained data models to train using a training data set. The selected untrained data models may include any type of untrained ML models for supervised, semi-supervised, self-supervised, or unsupervised machine learning. The selected untrained data models may be specified based upon input (e.g., user input) specifying relevant parameters to use as predicted variables or other variables to use as potential explanatory variables. For example, the selected untrained data models may be specified to generate an output (e.g., a prediction) based upon the input. Conditions for training the ML model from the selected untrained data models may likewiseWSGR Docket No.: 69686-701.601be selected, such as limits on the ML model complexity or limits on the ML model refinement past a certain point. The ML model may be trained (e.g., via a computer system such as a server) using the training data set. In some cases, a first subset of the training data set may be selected to train the ML model. The selected untrained data models may then be trained on the first subset of training data set using appropriate ML techniques, based upon the type of ML model selected and any conditions specified for training the ML model. In some cases, due to the processing power requirements of training the ML model, the selected untrained data models may be trained using additional computing resources (e.g., cloud computing resources). Such training may continue, in some cases, until at least one aspect of the ML model is validated and meets selection criteria to be used as a predictive model.

[0060] In some cases, one or more aspects of the ML model may be validated using a second subset of the training data set (e.g., distinct from the first subset of the training data set) to determine accuracy and robustness of the ML model. Such validation may include applying the ML model to the second subset of the training data set to make predictions derived from the second subset of the training data. The ML model may then be evaluated to determine whether performance is sufficient based upon the derived predictions. The sufficiency criteria applied to the ML model may vary depending upon the size of the training data set available for training, the performance of previous iterations of trained models, or user-specified performance requirements. If the ML model does not achieve sufficient performance, additional training may be performed. Additional training may include refinement of the ML model or retraining on a different first subset of the training dataset, after which the new ML model may again be validated and assessed. When the ML model has achieved sufficient performance, in some cases, the ML may be stored for present or future use. The ML model may be stored as sets of parameter values or weights for analysis of further input (e.g., further relevant parameters to use as further predicted variables, further explanatory variables, further user interaction data, etc.), which may also include analysis logic or indications of model validity in some instances. In some cases, a plurality of ML models may be stored for generating predictions under different sets of input data conditions. In some embodiments, the ML model may be stored in a database (e.g., associated with a server).Decision Tree and Random Forest

[0061] As described above, the machine learning model may implement a decision tree. A decision tree may be a supervised ML algorithm that can be applied to both regression and classification problems. Decision trees may mimic the decision-making process of a human brain. For example, a decision tree may grow from a root (base condition), and when it meetsWSGR Docket No.: 69686-701.601a condition (internal node / feature), it may split into multiple branches. The end of the branch that does not split anymore may be an outcome (leaf). A decision tree can be generated using a training data set according to the following operations: (1) Starting from a root node (the entire dataset), the algorithm may split the dataset in two branches using a decision rule or branching criterion; (2) each of these two branches may generate a new child node; (3) for each new child node, the branching process may be repeated until the dataset cannot be split any further; (4) each branching criterion may be chosen to maximize information gain (e.g., a quantification of how much a branching criterion reduces a quantification of how mixed the labels are in the children nodes). The labels may be the data or the classification that is predicted by the decision tree.

[0062] A random forest regression is an extension of the decision tree model that tends to yield more robust predictions by stretching the use of the training data partition. Whereas a decision tree may make a single pass through the data, a random forest regression may bootstrap 50% of the data (e.g., with replacement) and build many trees. Rather than using all explanatory variables as candidates for splitting, a random subset of candidate variables may be used for splitting, which may enable trees that have completely different data and different variables (hence the term random). The predictions from the trees, collectively referred to as the “forest,” may be then averaged together to produce the final prediction. Many trees (e.g., one hundred trees) may be included in a random forest model, with a number (e.g., 3, 6, 10, etc.) of terms sampled per split, a minimum of number (e.g., 1, 2, 4, 10, etc.) of splits per tree, and a minimum split size (e.g., 16, 32, 64, 128, 256, etc.). Random forests may be trained in a similar way as decision trees. Specifically, training a random forest may include the following operations: (1) select randomly k features from the total number of features; (2) create a decision tree from these k features using the same operations as for generating a decision tree; and (3) repeat the previous two operations until a target number of trees is created.Computer Vision

[0063] The systems, the methods, the computer-readable media, and the techniques disclosed herein may implement one or more computer vision techniques. Computer vision is a field of artificial intelligence that uses computers to interpret and understand the visual world at least in part by processing one or more digital images from cameras and videos. In some instances, computer vision may use deep learning models (e.g., convolutional neural networks). Bounding boxes may be used in object detection techniques within computer vision. Bounding boxes may be annotation markers drawn around objects in an image.WSGR Docket No.: 69686-701.601Bounding boxes, are often, although not always, may be rectangularly shaped. Bounding boxes may be applied by humans to training data sets. However, bounding boxes may also be applied to images by a trained machine learning that is trained to detect one or more different objects (e.g., humans, hands, faces, cars, etc.). In addition or in alternative to bounding boxes detection and tracking techniques may use any object detection annotation techniques, such as semantic segmentation, instance segmentation, polygon annotation, non-polygon annotation, landmarking, 3D cuboids, etc. In some embodiments, computer vision is used to analyze microscopy images, such as morphology of cells or organisms.Support Vector Machine

[0064] As also described above, the machine learning model may implement support vector machine learning techniques. In machine learning, support vector machines (SVMs) may be supervised learning models with associated learning algorithms that analyze data for classification and regression analysis. SVMs may be a robust prediction method, being based on statistical learning. SVMs may be well-suited for domains characterized by the existence of large amounts of data, noisy patterns, or the absence of general theories.

[0065] In general terms, SVMs may map input vectors into high dimensional feature space through non-linear mapping function, chosen a priori. In this high dimensional feature space, an optimal separating hyperplane may be constructed. The optimal hyperplane may then be used to determine things such as class separations, regression fit, or accuracy in density estimation. More formally, a SVM constructs a hyperplane or set of hyperplanes in a high or infinite-dimensional space, which can be used for classification, regression, or other tasks like outlier detection.

[0066] Support vectors may be defined as the data points that lie closest to the decision surface (or hyperplane). Support vectors may therefore be the data points that are most difficult to classify and may have direct bearing on the optimum location of the decision surface. Given a set of training examples, each marked as belonging to one of two categories, an SVM training algorithm may build a model that assigns new examples to one category or the other, making it a non-probabilistic binary linear classifier (although methods such as Platt scaling exist to use SVM in a probabilistic classification setting). SVM may map training examples to points in space so as to maximize the width of the gap between the two categories. New examples may then be mapped into that same space and predicted to belong to a category based on which side of the gap they fall. In addition to performing linear classification, SVMs can efficiently perform a non-linear classification using what is called the kernel trick, implicitly mapping their inputs into high-dimensional feature spaces.WSGR Docket No.: 69686-701.601

[0067] Within a support vector machine, the dimensionally of the feature space may be large. For example, a fourth-degree polynomial mapping function may cause a 200-dimensional input space to be mapped into a 1.6 billionth dimensional feature space. The kernel trick and the Vapnik-Chervonenkis dimension may allow the SVM to thwart the “curse of dimensionality” limiting other methods and effectively derive generalizable answers from this very high dimensional feature space. Accordingly, SVMs may assist in discovering knowledge from vast amounts of input data.

[0068] Patent applications directed to support vector machines include, U.S. patent application Ser. Nos. 09 / 303,386; 09 / 303,387; 09 / 303,389; 09 / 305,345; all filed May 1, 1999; and U.S. patent application Ser. No. 09 / 568,301, filed May 9, 2000; and U.S. patent application Ser. No. 09 / 578,011, filed May 24, 2000 and also claims the benefit of U.S.Provisional Patent Application No. 60 / 161,806, filed Oct. 27, 1999; of U.S. Provisional Patent Application No. 60 / 168,703, filed Dec. 2, 1999; of U.S. Provisional Patent Application No. 60 / 184,596, filed Feb. 24, 2000; and of U.S. Provisional Patent Application Ser. No. 60 / 191,219, filed Mar. 22, 2000; all of which are herein incorporated in their entireties.LSTM

[0069] Long short-term memory (LSTM) may be an artificial neural network used in the fields of artificial intelligence and deep learning. Unlike standard feedforward neural networks, LSTM may use feedback connections. The LSTM architecture may provide a short-term memory for a recurrent neural network (RNN). Such RNN can process not only single data points (such as images), but also entire sequences of data (such as speech or video). This characteristic may mean that LSTM networks are well-suited for processing and predicting data. The name of LSTM may refer to the analogy that a standard RNN has both “long-term memory” and “short-term memory.” The connection weights and biases in the RNN may change once per episode of training, analogous to how physiological changes in synaptic strengths store long-term memories; the activation patterns in the network may change once per time-step, analogous to how the moment-to-moment change in electric firing patterns in the brain store short-term memories. The LSTM architecture may provide a shortterm memory for an RNN that can last many (e.g., thousands) timesteps.

[0070] In some cases, a LSTM unit may comprise a cell, an input gate, an output gate, and a forget gate. The cell may remember values over arbitrary time intervals and the input gate, the output gate, and the forget gate may regulate the flow of information into and out of the cell. Forget gates may be used to decide what information to discard from a previous state by assigning a previous state, compared to a current input, a value between 0 and 1 (e.g., aWSGR Docket No.: 69686-701.601(rounded) value of 1 may mean to keep the information, and a value of 0 means to discard it). The input gate may decide which pieces of new information to store in the current state, using the same system as the forget gates. The output gate may control which pieces of information in the current state to output (e.g., by assigning a value from 0 to 1 to the information, considering the previous and current states). Selectively outputting relevant information from the current state may allow the LSTM network to maintain useful, long-term dependencies to make predictions, both in current and future time-steps. LSTM networks may be well-suited to classifying, processing and making predictions based on time series data, since there can be lags of unknown duration between important events in a time series. LSTMs may resolve the vanishing gradient problem that can be encountered when training traditional RNNs. Relative insensitivity to gap length may be an advantage of LSTM over RNNs, hidden Markov models and other sequence learning methods in numerous applications.

[0071] In some cases, LSTMs may be used with one or more various types of neural networks (e.g., convolutional neural networks (CNNs), deep neural network (DNNs), recurrent neural networks (RNNs), etc.). In some cases, CNNs, LSTM, and DNNs are complementary in their modeling capabilities and may be combined a unified architecture. For example, in such unified architecture, CNNs may be well-suited at reducing frequency variations, LSTMs may be well-suited at temporal modeling, and DNNs may be well-suited for mapping features to a more separable space. For example, input features to a ML model using LSTM techniques in the unified architecture may include segment features for each of a plurality of segments. To process the input features for each of the plurality of segments, the segment features for the segment may be processed using one or more CNN layers to generate first features for the segment; the first features may be processed using one or more LSTM layers to generate second features for the segment; and the second features may be processed using one or more fully connected neural network layers to generate third features for the segments, where the third features may be used for classification operations. In some examples, to process the first features using the one or more LSTM layers to generate the second features, the first features may be processed using a linear layer to generate reduced features having a reduced dimension from a dimension of the first features; and the reduced features may be processed using the one or more LSTM layers to generate the second features. Short-term features having a first number of contextual frames may be generated based on the input features, where features generated using the one or more CNN layers may include long-term features having a second number of contextual frames that are more than the first number of contextual frames of the short-term features. In some cases, the one orWSGR Docket No.: 69686-701.601more CNN layers, the one or more LSTM layers, and the one or more fully connected neural network layers may have been jointly trained to determine trained values of parameters of the one or more CNN layers, the one or more LSTM layers, and the one or more fully connected neural network layers. In some cases, the input features may include log-mel features having multiple dimensions. The input features may include one or more contextual frames indicating a temporal context of a signal (e.g., input data). Advantageously, implementations for such unified architecture may leverage complementary advantages associated with each of a CNN, LSTM, and DNN. For example, convolutional layers may reduce spectral variation in input, which may help the modeling of LSTM layers. Having DNN layers after LSTM layers may help reduce variation in the hidden states of the LSTM layers. Training the unified architecture jointly may provide a better overall performance. Training in the unified architecture may also remove the need to have separate CNN, LSTM and DNN architectures, which may be expensive (e.g., in computational resource, in network traffic, in financial resources, in energy consumption, etc.). By adding multi-scale information into the unified architecture, information may be captured at different time scales.Vision Transformer

[0072] A vision transformer (ViT) is a transformer-like model that handles vision processing tasks. While CNNs use convolution, a “local” operation bounded to a small neighborhood of an image, ViTs use self-attention, a “global” operation, since the ViT draws information from the whole image. This allows the ViT to capture distant semantic relevances in an image effectively. Advantageously, ViTs may be well-suited catching long-term dependencies. In some cases, ViTs may be a competitive alternative to convolutional neural networks as ViTs may outperform the current state-of-the-art CNNs by almost four times in terms of computational efficiency and accuracy. ViTs may be well-suited to object detection, image segmentation, image classification, and action recognition. Moreover, ViTs may be applied in generative modeling and multi-model tasks, including visual grounding, visual-question answering, and visual reasoning. In some cases, ViTs may represent images as sequences, and class labels for the image are predicted, which allows models to learn image structure independently. Input images may be treated as a sequence of patches where every patch is flattened into a single vector by concatenating the channels of all pixels in a patch and then linearly projecting it to the desired input dimension. For example, a ViT architecture may include the following operations: (A) split an image into patches; (B) flatten the patches; (C) generate lower-dimensional linear embeddings from the flattened patches; (D) add positional embeddings; (E) provide the sequence as an input to a standard transformer encoder; (F)WSGR Docket No.: 69686-701.601pretrain a model with image labels (e.g., fully supervised on a huge dataset); and (G) finetune on the downstream dataset for image classification. In some cases, there may be multiple blocks in a ViT encoder, with each block comprising three major processing elements: (1) Layer Norm; (2) Multi-head Attention Network; and (3) Multi-Layer Perceptrons. The Layer Norm may keep the training process on track and enable the model adapt to the variations among the training images. The Multi-head Attention Network may be a network responsible for generating attention maps from the given embedded visual tokens. These attention maps may help the network focus on the most critical regions in the image, such as object(s). The Multi-Layer Perceptrons may be a two-layer classification network with a Gaussian Error Linear Unit at the end. The final Multi-Layer Perceptrons block may be used as an output of the transformer. An application of softmax on this output can provide classification labels (e.g., if the application is image classification).Masked Autoencoder

[0073] Masked autoencoders (MAE) are scalable self-supervised learners for computer vision. The MAE leverages the success of autoencoders for various imaging and natural language processing tasks. Some computer vision models may be trained using supervised learning, such as using humans to look at images and created labels for the images, so that the model could learn the patterns of those labels (e.g., a human annotator would assign a class label to an image or draw bounding boxes around objects in the image). In contrast, selfsupervised learning may not use any human-created labels. One technique for self-supervised image processing training using an MAE is for before an image is input into an encoder transformer, a certain set of masks are applied to the image. Due to the masks, pixels are removed from the image and therefore the model is provided an incomplete image. At a high level, the model’s task is to now learn what the full, original image looked like before the mask was applied.

[0074] In other words, MAE may include masking random patches of an input image and reconstructing the missing pixels. The MAE may be based on two core designs. First, an asymmetric encoder-decoder architecture, with an encoder that operates on the visible subset of patches (without mask tokens), along with a lightweight decoder that reconstructs the original image from the latent representation and mask tokens. Second, masking a high proportion of the input image, e.g., 75%, may yield a nontrivial and meaningful self-supervisory task. Coupling these two core designs enables training large models efficiently and effectively, thereby accelerating training (e.g., by 3* or more) and improving accuracy. MAE techniques may be scalable, enabling learning of high-capacity models that generalizeWSGR Docket No.: 69686-701.601well, e.g., a vanilla ViT-Huge model. As mentioned, the MAE may be effective in pretraining ViTs for natural image analysis. In some cases, the MAE uses the characteristic of redundancy of image information to observe partial images to reconstruct original images as a proxy task, and the encoder of the MAE may have the capability of deducing the content of the masked image area by aggregating context information. This contextual aggregation capability may be important in the field of image processing and analysis.III. Synthetic Genomes and Biological entity

[0075] Provided herein are methods of synthesizing synthetic genomes. In some instances, synthetic genomes encode for a biological entity (including but not limited to a living organism, virus, or other biological entity). In some instances, synthetic genomes are introduced into cells to create fully synthetic organisms and synthetic biological entities, which have never existed in nature. In some instances a synthetic genome is at least 1, 5, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100 MB or more in size. In some instances, the synthetic genome comprises no more than 10, 20, 30, 40, 50, 60, 70, or no more than 80% sequence homology with any existing genome. In other embodiments, the synthetic genome comprises no more than 5, 10, 15, 20, 25, 30, 40, 50, or no more than 60% of its genes or non-coding regions derived from any single naturally occurring organism.

[0076] In some instances, an organism described herein comprises a prokaryote or eukaryote. In some instances, an organism described herein comprises a fungi, plant, animal, virus, or microbe. In some instances, an organism described herein does not comprise a human organism.

[0077] In some instances, the methods disclosed herein are configured to generate a synthetic bacterium. In some instances, the synthetic bacterium differs from all naturally occurring bacteria by one or more of: Cell Morphology (Differences in size, shape (e.g., round, oval, elongated), and budding pattern (e.g., multilateral vs. unipolar); Colony Morphology (Variations in color, texture (smooth, wrinkled, mucoid), margin, and elevation on solid media; Reproduction (Differences in asexual (budding vs. fission) and sexual reproduction (formation of ascospores or basidiospores); Metabolism (Variation in fermentation and respiration capabilities, including sugar utilization (e.g., glucose, maltose, galactose) and ethanol production; Growth Conditions (Differences in optimal temperature, pH tolerance, salt tolerance, and ability to grow under anaerobic vs. aerobic conditions; Stress Resistance (Ability to survive in extreme environments (e.g., high osmolarity, desiccation, oxidative stress, antifungal resistance); Pathogenicity (Differences in virulence, ability to colonize hosts, and potential to cause disease (e.g., Candida albicans vs. Saccharomyces cerevisiae);WSGR Docket No.: 69686-701.601Cell Wall Composition (Differences in polysaccharide composition affecting immune recognition and antifungal susceptibility; Secondary Metabolites (Production of unique bioactive compounds such as toxins (e.g., killer toxins) or flavor compounds in brewing and baking.

[0078] In some instances, the methods disclosed herein are configured to generate a synthetic virus. In some instances, the synthetic bacteriophage differs from all naturally occurring viruses by one or more of: Morphology (Differences in shape (e.g., icosahedral, helical, complex), size, and presence or absence of an envelope; Genome Type (Variations in nucleic acid composition (DNA vs. RNA), strandedness (single-stranded vs. double-stranded), and polarity (+ or (sense for RNA viruses); Envelope Presence (Whether the virus has a lipid envelope or is non-enveloped (naked), influencing stability and transmission; Replication Cycle (Differences in host cell entry (receptor usage), replication strategy (nuclear vs. cytoplasmic), and genome integration (e.g., retroviruses); Host Range (Variation in preferred host species and tissue tropism (e.g., respiratory, gastrointestinal, neural); Transmission Mode (Differences in how the virus spreads (e.g., airborne, fecal -oral, vector-borne, direct contact); Cytopathic Effects (Distinct effects on host cells, such as syncytia formation, apoptosis, or inclusion body formation; Immune Evasion (Differences in evasion strategies, such as antigenic variation, latency, or immune suppression; Pathogenicity (Variation in virulence, symptoms, and disease severity in infected hosts. In some instances a phage library is configured for therapeutic use (e.g., treatment of bacterial infections).

[0079] In some instances, the methods disclosed herein are configured to generate a synthetic yeast. In some instances, the synthetic yeast differs from all naturally occurring yeast by one or more of: Cell Morphology (Differences in size, shape (e.g., round, oval, elongated), and budding pattern (e.g., multilateral vs. unipolar); Colony Morphology (Variations in color, texture (smooth, wrinkled, mucoid), margin, and elevation on solid media; Reproduction (Differences in asexual (budding vs. fission) and sexual reproduction (formation of ascospores or basidiospores); Metabolism (Variation in fermentation and respiration capabilities, including sugar utilization (e.g., glucose, maltose, galactose) and ethanol production; Growth Conditions (Differences in optimal temperature, pH tolerance, salt tolerance, and ability to grow under anaerobic vs. aerobic conditions; Stress Resistance (Ability to survive in extreme environments (e.g., high osmolarity, desiccation, oxidative stress, antifungal resistance); Pathogenicity (Differences in virulence, ability to colonize hosts, and potential to cause disease (e.g., Candida albicans vs. Saccharomyces cerevisiae); Cell Wall Composition (Differences in polysaccharide composition affecting immuneWSGR Docket No.: 69686-701.601recognition and antifungal susceptibility; Secondary Metabolites (Production of unique bioactive compounds such as toxins (e.g., killer toxins) or flavor compounds in brewing and baking.

[0080] Synthetic genomes, once synthesized, may be introduced into cells using any know methods in the art (CRISPR / cas9, recombinases, etc.). In some instances, portions of a synthetic genome are introduced iteratively using a positive / negative selection system such as GENESIS (NatProtoc. 2021 May;16(5):2345-2380).IV. Kits

[0081] In some embodiments, the kit comprises the compositions described herein, which can be used to perform the methods described herein. Kits comprise an assemblage of materials or components, including at least one of the compositions. In some embodiments, the kits contains all of the components necessary and / or sufficient to create a polynucleotide library or assemble a nucleic acid sequence, including all controls, directions for performing assays, and any necessary software for analysis and presentation of results. In some embodiments, the kit comprises a nick repair enzyme, a ligase, a buffer, or a combination thereof. In some embodiments, the buffer has a pH of 5-9.

[0082] Instructions for use may be included in the kit. Optionally, the kit also contains other useful components, such as, diluents, buffers, pharmaceutically acceptable carriers, syringes, catheters, applicators, pipetting or measuring tools, bandaging materials or other useful paraphernalia. The materials or components assembled in the kit can be provided to the practitioner stored in any convenient and suitable ways that preserve their operability and utility. For example, the components can be in dissolved, dehydrated, or lyophilized form; they can be provided at room, refrigerated or frozen temperatures. The components are typically contained in suitable packaging material(s). As employed herein, the phrase “packaging material” refers to one or more physical structures used to house the contents of the kit, such as compositions and the like. The packaging material is constructed by well-known methods, preferably to provide a sterile, contaminant-free environment. The packaging materials employed in the kit are those customarily utilized in gene expression assays and in the administration of treatments. As used herein, the term “package” refers to a suitable solid matrix or material such as glass, plastic, paper, foil, and the like, capable of holding the individual kit components. Thus, for example, a package can be a glass vial or prefilled syringes used to contain suitable quantities of the pharmaceutical composition. The packaging material has an external label which indicates the contents and / or purpose of the kit and its components.WSGR Docket No.: 69686-701.601V. Definitions

[0083] Throughout this disclosure, various embodiments are presented in a range format. It should be understood that the description in range format is merely for convenience and brevity and should not be construed as an inflexible limitation on the scope of any embodiments. Accordingly, the description of a range should be considered to have specifically disclosed all the possible subranges as well as individual numerical values within that range to the tenth of the unit of the lower limit unless the context clearly dictates otherwise. For example, description of a range such as from 1 to 6 should be considered to have specifically disclosed subranges such as from 1 to 3, from 1 to 4, from 1 to 5, from 2 to 4, from 2 to 6, from 3 to 6 etc., as well as individual values within that range, for example, 1.1, 2, 2.3, 5, and 5.9. This applies regardless of the breadth of the range. The upper and lower limits of these intervening ranges may independently be included in the smaller ranges, and are also encompassed within the invention, subject to any specifically excluded limit in the stated range. Where the stated range includes one or both of the limits, ranges excluding either or both of those included limits are also included in the invention, unless the context clearly dictates otherwise.

[0084] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of any embodiment. As used herein, the singular forms “a,” “an” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms “comprises” and / or “comprising,” when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. As used herein, the term “and / or” includes any and all combinations of one or more of the associated listed items.

[0085] Unless specifically stated or obvious from context, as used herein, the term “nucleic acid” encompasses double- or triple-stranded nucleic acids, as well as single-stranded molecules. In double- or triple-stranded nucleic acids, the nucleic acid strands need not be coextensive (i.e., a double-stranded nucleic acid need not be double-stranded along the entire length of both strands). Nucleic acid sequences, when provided, are listed in the 5’ to 3’ direction, unless stated otherwise. Methods described herein provide for the generation of isolated nucleic acids. Methods described herein additionally provide for the generation of isolated and purified nucleic acids. A “nucleic acid” as referred to herein can comprise atWSGR Docket No.: 69686-701.601least 5, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 125, 150, 175, 200, 225, 250, 275, 300, 325, 350, 375, 400, 425, 450, 475, 500, 600, 700, 800, 900, 1000, 1100, 1200, 1300, 1400, 1500, 1600, 1700, 1800, 1900, 2000, or more bases in length. Moreover, provided herein are methods for the synthesis of any number of polypeptide-segments encoding nucleotide sequences, including sequences encoding non-ribosomal peptides (NRPs), sequences encoding non-ribosomal peptide-synthetase (NRPS) modules and synthetic variants, polypeptide segments of other modular proteins, such as antibodies, polypeptide segments from other protein families, including non-coding DNA or RNA, such as regulatory sequences e.g. promoters, transcription factors, enhancers, siRNA, shRNA, RNAi, miRNA, small nucleolar RNA derived from microRNA, or any functional or structural DNA or RNA unit of interest. The following are non-limiting examples of polynucleotides: coding or noncoding regions of a gene or gene fragment, intergenic DNA, loci (locus) defined from linkage analysis, exons, introns, messenger RNA (mRNA), transfer RNA, ribosomal RNA, short interfering RNA (siRNA), short-hairpin RNA (shRNA), micro-RNA (miRNA), small nucleolar RNA, ribozymes, complementary DNA (cDNA), which is a DNA representation of mRNA, usually obtained by reverse transcription of messenger RNA (mRNA) or by amplification; DNA molecules produced synthetically or by amplification, genomic DNA, recombinant polynucleotides, branched polynucleotides, plasmids, vectors, isolated DNA of any sequence, isolated RNA of any sequence, nucleic acid probes, and primers. cDNA encoding for a gene or gene fragment referred to herein may comprise at least one region encoding for exon sequences without an intervening intron sequence in the genomic equivalent sequence.

[0086] In some embodiments, provided compositions and methods provided herein are designed to synthesize one or more oligonucleotides that are or act as aptamers. Aptamers are nucleic acid or peptide molecules that bind to a particular molecule of interest with high affinity and specificity. Generally, aptamers are engineered through repeated rounds of in vitro selection or equivalently, SELEX (systematic evolution of ligands by exponential enrichment) to bind to various molecular targets such as small molecules, proteins, nucleic acids, and even cells, tissues and organisms. Provided herein are Al-assisted methods for generation of aptamers having specific properties such as binding affinity or catalytic activity.

[0087] Unless specifically stated or obvious from context, as used herein, the term “about” in reference to a number or range of numbers is understood to mean the stated number andWSGR Docket No.: 69686-701.601numbers + / - 10% thereof, or 10% below the lower listed limit and 10% above the higher listed limit for the values listed for a range.

[0088] Primers referred to in the exemplary workflows mentioned herein as “universal primers,” are short polynucleotides that recognize a primer binding site common to multiple DNA fragments. However, these workflows are not limited to only use of universal primers, and fragment-specific primers may be incorporated in addition or alternatively. In addition, while exemplary workflows described herein refer to assembly of gene fragments, they are not limited as such and are applicable to the assembly of longer nucleic acids in general. As used in this specification and the appended claims, the terms “artificial intelligence,” “artificial intelligence techniques,” “artificial intelligence operation,” and “artificial intelligence algorithm” generally refer to any system or computational procedure that may take one or more actions that simulate human intelligence processes for enhancing or maximizing a chance of achieving a goal. The term “artificial intelligence” may include “generative modeling,” “machine learning” (ML), or “reinforcement learning” (RL).

[0001] As used in this specification and the appended claims, the terms “machine learning,” “machine learning techniques,” “machine learning operation,” and “machine learning model” generally refer to any system or analytical or statistical procedure that may progressively improve computer performance of a task. In some cases, ML may generally involve identifying and recognizing patterns in existing data in order to facilitate making predictions for subsequent data. ML may include a ML model (which may include, for example, a ML algorithm). Machine learning, whether analytical or statistical in nature, may provide deductive or abductive inference based on real or simulated data. The ML model may be a trained model. ML techniques may comprise one or more supervised, semi-supervised, selfsupervised, or unsupervised ML techniques. For example, an ML model may be a trained model that is trained through supervised learning (e.g., various parameters are determined as weights or scaling factors). ML may comprise one or more of regression analysis, regularization, classification, dimensionality reduction, ensemble learning, meta learning, association rule learning, cluster analysis, anomaly detection, deep learning, or ultra-deep learning. ML may comprise, but is not limited to: k-means, k-means clustering, k-nearest neighbors, learning vector quantization, linear regression, non-linear regression, least squares regression, partial least squares regression, logistic regression, stepwise regression, multivariate adaptive regression splines, ridge regression, principal component regression, least absolute shrinkage and selection operation (LASSO), least angle regression, canonical correlation analysis, factor analysis, independent component analysis, linear discriminantWSGR Docket No.: 69686-701.601analysis, multidimensional scaling, non-negative matrix factorization, principal components analysis, principal coordinates analysis, projection pursuit, Sammon mapping, t-distributed stochastic neighbor embedding, AdaBoosting, boosting, gradient boosting, bootstrap aggregation, ensemble averaging, decision trees, conditional decision trees, boosted decision trees, gradient boosted decision trees, random forests, stacked generalization, Bayesian networks, Bayesian belief networks, naive Bayes, Gaussian naive Bayes, multinomial naive Bayes, hidden Markov models, hierarchical hidden Markov models, support vector machines, encoders, decoders, auto-encoders, stacked auto-encoders, perceptrons, multi-layer perceptrons, artificial neural networks, feedforward neural networks, convolutional neural networks, recurrent neural networks, long short-term memory, deep belief networks, deep Boltzmann machines, deep convolutional neural networks, deep recurrent neural networks, or generative adversarial networks.EXAMPLES

[0089] The following examples are given for the purpose of illustrating various embodiments of the invention and are not meant to limit the present invention in any fashion. The present examples, along with the methods described herein are presently representative of preferred embodiments, are exemplary, and are not intended as limitations on the scope of the invention. Changes therein and other uses which are encompassed within the spirit of the invention as defined by the scope of the claims will occur to those skilled in the art.Example 1: Synthesizing highly repetitive proteins

[0090] Repetitive sequences often produce errors with providing the underlying cDNA template. For instance, a nucleic acid template encoding COL1 Al (type 1 collagen), ELN (tropoelastin), or rbcL (rubisco large subunit) is synthesized de novo. In some instances, the nucleic acid encodes cDNA of the protein. In other embodiments, the nucleic acid encodes for the entire gene sequence including introns and exons. The sequence is first computationally divided into fragments of approximately similar lengths to form an assembly library. For example, a nucleic acid template of 20kb could be divided into 100 fragments of approximately 200 bases each (allowing for potential 1-15% in length variation based on the design). These small fragments are obtained using standard array-based nucleotide synthesis chemistry. An exemplary arrangement of the assembly is shown in FIG. 4B and FIG. 5A. Each internal fragment has a top strand comprising a unique index region at both the 5’ and the 3’ ends of the fragment (40 bases in length), and each bottom strand has a stabilizing region at the 5’ end (15 bases in length). The fragment comprising the 5’ end of the nucleicWSGR Docket No.: 69686-701.601acid template has a top strand with a 3’ index region (40 bases in length) and a bottom strand with a 3’ stabilizing region (15 bases in length). The fragment comprising the 3’ end of the nucleic acid template has a top strand with a 5’ index region (40 bases in length) and a bottom strand with a 3’ stabilizing region (15 bases in length). In some instances, the methods include those described in Robinson, N.E., Zhang, W., Ghosh, R. et al. Construction of complex and diverse DNA sequences using DNA three-way junctions. Nature 651, 491-500 (2026), which is herein incorporated by reference.

[0091] The 40 base long index regions are computationally designed using artificial intelligence, and optionally may be 10-50 bases in length. The 15 base stabilizing regions are endogenous to the nucleic acid template, and optionally may be 10-30 bases in length. Each index region anneals to a complementary index region of an adjacent fragment, and the fragments ligate at the stabilizing region in a specific order as specified by the design - this prevents fragments from ligating in an erroneous order despite each fragment comprising highly repetitive sequences. The ligated strand is amplified to produce the full-length nucleic acid template without any index regions. In general, the method results in higher assembly yield and accuracy than other comparator assembly methods (e.g., FIG. 3A-3B, 4A) because it relies primarily on non-endogenous hybridization regions for assembly vs. endogenous regions (e.g., PCA, Gibson assembly, etc.). The template may then be further amplified for gel electrophoresis size analysis, sequencing, or expression in an organism via a vector.Example 2: Synthesizing proteins from GC rich templates

[0092] GC-rich nucleic acid templates encoding proteins often produce errors during synthesis due to unwanted hybridization. For instance, a nucleic acid template encoding cellulase, FGF2 (Fibroblast growth factor 2), or TERT (component of telomerase) is synthesized de novo using the general methods of Example 1. In an alternative method, the template is assembled using PCA or Gibson assembly as described in FIGS. 3A-3B. In other embodiments, a GC rich template having a length of at least 5kb and at least 75% GC content is assembled. In some instances the nucleic acid template contains cDNA, or contains the entire gene (including non-coding / regulatory regions, such as introns and exons). In other embodiments, the template comprises one or more sections of 10-500 bases which are greater than 60% GC content.WSGR Docket No.: 69686-701.601Example 3: GWAS deconvolution

[0093] A genome wide association study is performed to identify a causal variant associated with a phenotypic trait in an organism. The genotype of at least 10,000, 25,000, 50,000, 75,000, 100,000 or more than 100,000 people is obtained and phenotypic traits are measured. Traits are associated with genetic variants, and the majority of such genetic variations occur in non-coding regions. However, the genomic regions associated with specific traits may span thousands to millions of bases. Multiple variants may also be associated with a particular phenotype. Using the general methods of Example 1, sequences including the individual variants are synthesized and tested in in-vivo models to identify causal variants. In an alternative method, the sequences are assembled using PCA or Gibson assembly as described in FIGS. 3A-3BExample 4: Artificial intelligence-assisted protein function

[0094] An existing protein’s properties are used as a baseline. Artificial intelligence and machine learning as described herein is used to evolve new protein function (e.g., higher stability, enzymatic activity, expanded substrate scope, efficiency, etc.) by utilizing a dataset of existing protein properties and structure. The general methods of Example 1 are used to assemble a library of sequences encoding for the potential new protein. The proteins are expressed and the new function is tested in vitro or in vivo. A protein with a new function is identified, and the processes is optionally iterated to optimize the new function.

[0095] In an alternative embodiment, the libraries are assembled using PCA or Gibson assembly as described in FIGS. 3A-3B.Example 5: Artificial intelligence-assisted synthetic protein design

[0096] A database of 2000 or more proteins of known structure are used. Machine learning methods are used to arrive at a synthetic protein sequence with predicted secondary and ternary structure when folded. The nucleic acid sequence encoding the protein or the protein has no more than 50%, 60%, 70%, 80%, or 90% sequence identity with any known protein. The protein is then synthesized and the functions of the protein tested using the general methods of Example 1 and Example 4. In some instances, the protein comprises non-canonical amino acids or post-translational modifications. In an alternative embodiment, the libraries are assembled using PCA or Gibson assembly as described in FIGS. 3A-3B.WSGR Docket No.: 69686-701.601

[0097] In other embodiments, a similar approach is used to generate aptamers, such as those with new biological functions (catalysis, binding, etc.). Such molecules in some instances are used for diagnostics and / or therapeutics.Example 6: Designing and optimizing gene circuits

[0098] A gene circuit for production of at least one target molecule is designed. The gene circuit contains at least one promoter and one gene expression product. The gene circuit in some embodiments also comprises a terminator. The target molecule may be a protein. The target molecule may be a RNA, such as a long noncoding RNA (IncRNA). The gene circuit may express multiple intermediate products. For instance, the gene circuit may express multiple enzymes needed to produce a metabolic product.

[0099] In some instance, the gene circuit has at least one feedback mechanism. The gene circuit may regulate expression of one or more intermediate products based on timing or levels of a different intermediate product. In some instances gene circuit encodes for one or more transcription factors which modulate expression of one or more elements of the gene circuit. Additional gene circuit elements such as gene switches, logic gates, and oscillators may also be employed; for general examples see Brophy et al., Principles of genetic circuit design (2014), Nature Methods 11, 508-520.

[0100] The gene circuit is assembled using the general nucleic acid assembly mechanisms described in Example 1. In some cases, the gene circuit is introduced into a bacterial cell, a yeast cell, or a mammalian cell using the methods described herein. In other instances, the gene circuit is expressed in vivo. The target molecule is produced. In an alternative embodiment, the libraries are assembled using PCA or Gibson assembly as described in FIGS. 3A-3BExample 7: Design and synthesis of vectors

[0101] A vector is redesigned to be optimized for delivery. The vector may be a lentiviral vector, an AAV vector or an HIV vector. The vector is optimized using the methods described herein to optimize a property of interest. In some instances, the vector is redesigned for reduced immunogenicity, enable blood-brain barrier, increase payload size, or increase transfection efficiency. The optimized vector is assembled using the general methods of Example 1. Vectors in some instances range from 5kb to 2000kb. Vectors designed herein in some instances are used to delivery large DNA templates synthesized using the general methods of Example 1 to an organism. In some instances, the vector comprises a virus. In anWSGR Docket No.: 69686-701.601alternative embodiment, the vector is assembled using PCA or Gibson assembly as described in FIGS.3A-3BExample 8: Design and product of gene therapies and gene editing

[0102] In some instances gene therapy may include gene editing. Gene editing may include epigenetic editing or transposon-mediated editing.

[0103] The gene therapy constructs are assembled using the general methods of Example 1. Once edited, the gene therapy constructs are transformed into appropriate vectors (viral or non-viral) as described herein and tested in mammalian cells. In a non-limiting example, large genomic deletions may cause disorders such as myelodysplastic syndrome (MDS), which is frequently associated with a deletion (del7q) in long arm of chromosome 7 (chr7q). The general methods of Example 1 are used to synthesize a 12Mb chromosomal section, and gene editing methods are used to re-introduce this missing section to otherwise diseased cells to restore normal cell function. In an alternative embodiment, the libraries are assembled using PCA or Gibson assembly as described in FIGS.3A-3B.Example 9: Designing and synthesizing a bacteriophage library

[0104] Artificial intelligence such as StripedHyena or megaDNA is used to design a library of synthetic bacteriophages (see the general methods described in Nguyen, et al. Science 386, 746 (2024) and Shao et al. Nat. Comm. (2024) 15:9392). The model is trained on a database of at least 1,000, 10,000, 100,000, 1,000,000 or more bacteriophage genomes. The resulting bacteriophage have a genome with at least 30%, 40%, 50%, 60%, 70% or 80% GC content. The bacteriophage genomes are synthesized de novo using the general methods of Example 1. Synthetic genomes may be filtered in-silico for the presence of essential genes prior to testing. The genomes are combined with proteins and packaged into bacteriophage particles. The bacteriophage particles are combined and tested against a panel of bacteria to identify potential targets. The bacteria may be antibiotic resistant. In an alternative embodiment, the libraries are assembled using PCA or Gibson assembly as described in FIGS.3A-3B.

[0105] The identified bacteriophage may be used as a direct therapeutic, for disinfecting surfaces (such as in a hospital), for food safety, or for other uses. In a non-limiting example, a library of phages infecting Salmonella enterica are used as the genomic data set for synthetic phage design (see Gao et al., BMC Genomics. 2020 May 26; 21:374 for an example phage library) Promoters, terminators, other regulatory elements, and coding regions of the phage genomes are organized and shuffled for recoding into fully synthetic phage genomes. TheWSGR Docket No.: 69686-701.601fully synthetic phage genomes are then tested for infectivity of Salmonella enterica, and those with the highest antibacterial properties may be iteratively improved with additional rounds of design and selection.Example 10: Designing and synthesizing a synthetic genome

[0106] A system comprising an artificial intelligence module employing StripedHyena or megaDNA is used to design a synthetic genome. The model is trained on a database of at least 1,000, 10,000, 100,000, 1,000,000 or more bacteriophage and prokaryotic genomes.

[0107] The system produces a sequence comprising a synthetic genome encoding a new species that has never previously existed in nature, and that is not directly based upon modified versions of existing genomes. The synthetic genome comprises an arrangement of genes and regulatory systems. The genome is at least 1, 5, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100 MB or more in size. The genome is synthesized and assembled using the general methods of Example 1. In an alternative embodiment, the genome is assembled using PCA or Gibson assembly as described in FIGS. 3A-3B.

[0108] The synthetic genome may comprise no more than 50% sequence homology with any existing genome. In other embodiments, the synthetic genome comprises no more than 25% of its genes or non-coding regions derived from any single naturally occurring organism.

[0109] Optionally, the genome is inserted into a cell to create an artificial organism.Synthetic genomes may be filtered in-silico for the presence of essential genes or other properties prior to testing in living organisms. Replacement of large genomic DNA sections in some instances is accomplished using a positive / negative selection method, such as Replicon Excision Enhanced Recombination (REXER) method and Genome Stepwise Interchange Synthesis (GENESIS), see the general methods described in Fredens, J. et al. Nature 569, 514-518 (2019) and Robertson, et al. Nat Protoc. 2021 May;16(5):2345-2380.

[0110] In other embodiments, the synthetic genome comprises one or more post-transcriptional modifications such as 5-hydroxymethylcytosine (5hmC), 5-methylcytosine (5mC), 5-formylcytosine (5fC), 5-carboxylcytosine (5caC), and 6-methyladenine (6mA).

[0111] In other embodiments, the synthetic genome comprises information encoding for an extinct species. In other embodiments, a synthetic genome comprises a recoded genome (e.g., having one or more codons reassigned, deleted, or added; see Fredens et al., above).Example 11: Assembly using 3-way junctions

[0112] Examples 1-10 are performed with modification - the general methods described for 3-way junction assembly as described in Robinson, N.E., Zhang, W., Ghosh, R. etWSGR Docket No.: 69686-701.601al. Construction of complex and diverse DNA sequences using DNA three-way junctions. Nature 651, 491-500 (2026) are used for assembly of fragments.

Claims

WSGR Docket No.: 69686-701.601CLAIMS WHAT IS CLAIMED IS:

1. A composition comprising a polynucleotide library, wherein the polynucleotide library comprises:(a) a first polynucleotide, whereinthe first polynucleotide comprises a first strand and a second strand, the first strand comprises a first fragment region, a first stabilizing region, and a first index region;the second strand comprises a second fragment region, wherein the first fragment region and the second fragment region are at least partially complementary; and(b) a second polynucleotide, whereinthe second polynucleotide comprises a third strand and a fourth strand, whereinthe third strand comprises a second index region and a third fragment region, andthe fourth strand comprises a second stabilizing region and a fourth fragment region;wherein the third fragment region and the fourth fragment region are at least partially complementary; andwherein the first index region and the second index region are at least partially complementary, and the first stabilizing region and the second stabilizing region are at least partially complementary;wherein the polynucleotides of the library collectively comprise one or more complex regions, and the polynucleotide library collectively encodes for a synthetic protein, aptamer, synthetic genome, synthetic gene circuit, or synthetic vector.

2. The composition of claim 1, wherein the one or more complex regions individually or collectively comprise secondary structure (optionally alpha helix, beta sheet, stemloop, pseudoknot, homodimer, heterodimer, hairpin loop, interior loop, bulge, or multiloop), inversions, palindromic sequences, repeats, high GC, or low AT.

3. The composition of claim 1 or 2, wherein the one or more of the fragment regions individually or collectively comprise at least two complex regions.

4. The composition of any one of claims 1-3, wherein one or more of the first fragment region, the second fragment region, the third fragment region, and the fourth fragmentWSGR Docket No.: 69686-701.601region is independently 100-1000, 100-2500, 500-5000, 500-10,000, or 500-100,000 nucleotides in length.

5. The composition of any one of claims 1-4, wherein one or more of the index sequences is independently 10-200 nucleotides in length.

6. The composition of any one of claims 1-5, wherein one or more of the stabilizing sequences is independently 10-200 nucleotides in length.

7. The composition of any one of the preceding claims, wherein the first stabilizing region and the second stabilizing region have a Tmof 40-95 degrees Celsius.

8. The composition of any one of the preceding claims, wherein the composition further comprises a nick repair enzyme.

9. The composition of any one of the preceding claims, wherein the composition further comprises a ligase.

10. The composition of any one of the preceding claims, wherein the composition further comprises a buffer.

11. The composition of any one of the preceding claims, wherein the composition further comprises a buffer having a pH of 5-9.

12. The composition of any one of the preceding claims, wherein the first index region and the second index region have a Tmof 40-95 degrees Celsius.

13. The composition of any one of the preceding claims, wherein the first index region and the second index region are less than 90, 95%, 97%, or 99% identical or complementary to any fragment region in the polynucleotide library.

14. The composition of any one of the preceding claims, wherein the first index region and the second index region are different than any fragment region in the polynucleotide library.

15. The composition of any one of the preceding claims, wherein any index region in the library comprises a Hamming or Levenshtein distance of at least two from any other index region.

16. The composition of any one of the preceding claims, wherein the second stabilizing region and the second fragment region are covalently attached.

17. The composition of any one of the preceding claims, wherein the first strand comprises from 5’ to 3’: the first fragment region, the first stabilizing region, and the first index region.

18. The composition of any one of the preceding claims, wherein the third strand comprises from 5’ to 3’ : the second index region and the third fragment region.WSGR Docket No.: 69686-701.60119. The composition of any one of the preceding claims, wherein the fourth strand comprises from 5’ to 3’: the fourth fragment region and the second stabilizing region.

20. The composition of any one of the preceding claims, wherein the first strand further comprises a 5’ third index sequence.

21. The composition of any one of the preceding claims, wherein the second strand further comprises a third 5’ stabilizing sequence.

22. The composition of any one of the preceding claims, wherein the first strand comprises a 3’ overhang.

23. The composition of claim , wherein the 3’ overhang comprises the first stabilizing region and the first index region.

24. The composition of any one of the preceding claims, wherein the third strand comprises a 5’ overhang.

25. The composition of claim , wherein the 5’ overhang comprises the second index region.

26. The composition of any one of the preceding claims, wherein the second nucleic acid comprises at least one non- complementary region.

27. The composition of claim 26, wherein the at least one non-complementary region is formed between the second index sequence and the second stabilizing region.

28. The composition of any one of the preceding claims, wherein the one or more of the fragment regions individually or collectively comprise at least 1, 2, 5, 10, 20, 30, 40, 50, 60, 75, 80, or at least 90% complex regions.

29. The composition of any one of the preceding claims, wherein the composition does not comprise one or more of a polymerase, exonuclease, endonuclease, SSB, or recombinase.

30. The composition of any one of the preceding claims, wherein the 5’ end of the first fragment region and the 3’ end of the third fragment region are not substantially complementary.

31. The composition of any one of the preceding claims, wherein the library comprises at least 5, 10, 15, 20, 30, 40, 50, 75, 100, or at least 500 unique polynucleotides.

32. A kit comprising the composition of any one of claims 1-31.

33. A method of nucleic acid assembly comprising:assembling two or more polynucleotide fragments into the nucleic acid, wherein the nucleic comprises one or more of complex regions.WSGR Docket No.: 69686-701.60134. The method of claim 33, wherein the one or more complex regions individually or collectively comprise secondary structure (optionally alpha helix, beta sheet, stemloop, pseudoknot, homodimer, heterodimer, hairpin loop, interior loop, bulge, or multiloop), inversions, palindromic sequences, repeats, high GC, or low AT.

35. The method of claim 33, wherein the nucleic acid comprises at least two complex regions.

36. The method of claim 33, wherein the nucleic acid comprises at least 1, 2, 5, 10, 20, 30, or 40% complex regions.

37. The method of any one of claims 33-36, wherein the method assembles at least 5, 10, 15, 20, 25, 30, 40, 50 or at least 100 polynucleotide fragments.

38. The method of any one of claims 33-37, wherein the nucleic acid is at least 1000, 3000, 5000, 10,000, 20,000, 50,000, 100,000, 500,000, 1 million, 10 million, 100 million, 1 billion, or at least 10 billion nucleotides in length.

39. The method of any one of claims 33-38, wherein the assembling comprises contacting two or more polynucleotide fragments with a nick repair enzyme.

40. The method of any one of claims 33-39, wherein the assembling comprises contacting two or more polynucleotide fragments with a ligase.

41. The method of any one of claims 33-40, wherein the nucleic acid originates from an animal, plant, bacteria, virus, fungi, or archaea.

42. The method of any one of claims 33-41, wherein the nucleic acid originates from a eukaryote.

43. The method of any one of claims 33-42, wherein the nucleic acid encodes a protein, a regulatory element, digital information, genome, or vector.

44. The method of any one of claims 33-43, wherein the nucleic acid encodes for a protein, wherein the protein optionally comprises collagen, elastin, rubisco, cellulase, fibroblast GF, apolipoprotein, an antibody, or telomerase.

45. The method of any one of claims 33-44, wherein the composition collectively encodes for an antibody library having a diversity of at least 106, 107, 108, 109, 1010, 1011or at least 1012unique members.

46. The method of any one of claims 33-45, wherein the nucleic acid encodes for a genome, wherein the genome comprises at least 10, 20, 30, 40, 50, 70, or at least 80% repetitive sequences.

47. The method of any one of claims 33-46, wherein the method further comprises amplifying the nucleic acid.WSGR Docket No.: 69686-701.60148. The method of any one of claims 33-47, wherein the method further comprises amplifying a single strand of the nucleic acid.

49. The method of any one of claims 33-48, wherein at least 50, 75, 90, 95, 98, or at least 99% of the amplicons have a length corresponding to the total length of all fragment regions and stabilizing regions in the composition.

50. The method of any one of claims 33-49, wherein at least 50, 75, 90, 95, 98, or at least 99% of the amplicons have a length corresponding to the total length of all fragment regions and stabilizing regions in the composition without purification.

51. The method of any one of claims 33-49, wherein at least 50, 75, 90, 95, 98, or at least 99% of the amplicons have a length corresponding to the total length of all fragment regions and stabilizing regions in the composition without digestion, filtration, size purification, or gel purification.

52. The method of any one of claims 33-51, wherein the method further comprises purifying the nucleic acid.

53. The method of any one of claims 33-52, wherein the polynucleotide fragments are synthesized.

54. The method of any one of claims 33-53, wherein the polynucleotide fragments are synthesized on a surface.

55. The method of any one of claims 33-54, wherein the polynucleotide fragments are synthesized using chemical synthesis or template-free enzymatic synthesis.

56. The method of any one of claims 33-55, wherein assembly occurs at 40-95 degrees Celsius.

57. The method of any one of claims 33-56, wherein assembly has a complexity score of at least 5, 10, 15, 20, 30, 50, or at least 100.

58. The method of claim 57, wherein assembly has a complexity score of at least 5, 10, 15, 20, 30, 50, or at least 100 when measured using an in-silico assembly tool.

59. The method of any one of claims 33-58, wherein the two or more polynucleotide fragments collectively encodes for a synthetic protein, aptamer, synthetic genome, synthetic gene circuit, or synthetic vector.

60. The method of any one of claims 33-59, wherein the nucleic acid is assembled in no more than 1 day, 1 week, 2 weeks, or 4 weeks.

61. The method of any one of claims 33-60, wherein the nucleic acid is assembled at a rate of at least 30kb, lOOkb, or 1Mb per week.WSGR Docket No.: 69686-701.60162. The method of any one of claims 33-61, wherein the method further comprises inserting the nucleic acid into a cell, genome, artificial chromosome, plasmid, or vector.

63. The method of claim 62, wherein inserting comprises recombination, TALEN, CRISPR, or Zn-finger endonucleases.

64. A system for nucleic acid assembly comprising:(a) an input comprising a nucleic acid sequence of nucleic acid to be assembled;and(b) one or more computer-implemented algorithms configured to:i. generate two or more fragments of the nucleic acid sequence;ii. append index sequences to a portion of the fragments; and iii. output a library of synthetic sequences, wherein a portion of the synthetic sequences comprise the index sequences.

65. The system of claim 64, wherein the system further comprises instructions to synthesize polynucleotide fragments corresponding to the synthetic sequences in the library.

66. The system of any one of claims 64-65, wherein the system further comprises a device configured to synthesize the polynucleotide fragments corresponding to the synthetic sequences in the library.

67. The system of any one of claims 64-66, wherein the system further comprises a device configured to assemble the polynucleotide fragments corresponding to the synthetic sequences in the library.

68. The system of any one of claims 64-67, wherein the device is configured to execute the method of any one of claims 33-63.

69. The system of claim 64, wherein the index sequences are less than 90, 95%, 97%, or 99% identical or complementary to any fragment region in the polynucleotide library.

70. The system of any one of claims 64-69, wherein each index sequences is different than any contiguous sequence in the nucleic acid sequence.

71. The system of any one of claims 64-70, wherein each index sequence has a Hamming or Levenshtein distance of at least two from any other index sequence.

72. The system of any one of claims 64-71, wherein the number of index sequences equals the number of fragments minus two.

73. The system of any one of claims 64-72, wherein the computer-implemented algorithms comprise use of machine learning.WSGR Docket No.: 69686-701.60174. The system of claim 73, wherein machine learning comprises inductive learning (e.g., decision / classification trees such as RF (random forest), C&RT, boosted trees, etc.), Probably Approximately Correct (PAC) learning, connectionist learning (e.g., neural networks (NN), artificial neural networks (ANN), neuro fuzzy networks (NFN), network structures, the Cox Proportional-Hazards Model (CPHM), perceptrons such as multi-layer perceptrons, multi-layer feed-forward networks, applications of neural networks, Bayesian learning in belief networks, reinforcement learning, support vector machines (e.g., Kernel methods), multivariate adaptive regression splines (MARS), Levenberg-Marquardt algorithms, Gauss-Newton algorithms, mixtures of Gaussians, gradient descent algorithms, or learning vector quantization (LVQ).

75. The system of any one of claims 64-74, wherein the library of synthetic sequences comprises at least 5, 10, 15, 20, 30, 40, 50, 75, 100, or at least 500 synthetic sequences.

76. A synthetic biological entity comprising a synthetic genome, wherein(a) the synthetic genomei. is at least 1, 5, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100 Mb or more in size;ii. comprises no more than 10, 20, 30, 40, 50, 60, 70, or no more than 80% sequence homology with any naturally occurring genome; or iii. comprises no more than 5, 10, 15, 20, 25, 30, 40, 50, or no more than 60% of its genes or non-coding regions derived from any single naturally occurring organism; and(b) the synthetic organism is distinguishable from naturally occurring organisms due to one or properties.

77. The synthetic biological entity of claim 76, wherein the organism is distinguishable from naturally occurring organisms due to at least 1, 2, 3, 5, 10, 20, or at least 30 properties.

78. The synthetic biological entity of claim 76, wherein the genome comprises coding or non-coding DNA regions from at least 5, 10, 15, 20, 25, 30 or at least 30 different organisms.

79. The synthetic biological entity of claim 76, wherein the organism comprises a bacterium, yeast, or virus.

80. The synthetic biological entity of claim 76, wherein the organism comprises a synthetic bacterium and the one or more properties are: cell morphology, colonyWSGR Docket No.: 69686-701.601morphology, reproduction, metabolism, growth conditions, stress resistance, pathogenicity, cell wall composition, or secondary metabolites.

81. The synthetic biological entity of claim 76, wherein the organism comprises a synthetic virus and the one or more properties are: morphology, genome type, envelope presence, replication cycle, host range, transmission mode, cytopathic effects, immune evasion, or pathogenicity.

82. The synthetic biological entity of claim 76, wherein the organism comprises a synthetic yeast and the one or more properties are: cell morphology; colony morphology reproduction, metabolism, growth conditions, stress resistance, pathogenicity, cell wall composition, or secondary metabolites.

83. A method of synthesizing the synthetic biological entity of any one of claims 76-82 comprising synthesizing the synthetic genome and incorporating the synthetic genome or iteratively incorporating portions thereof into a cell or other vector.