DNA fountain: As storage system

Some problems which is necessary to overcome

To read even a single byte of information from storage the entire DNA pool must be sequenced and decode. Which is very time consuming and costly. So, some researchers proposed method for random access that uses PCR to amplify only the desired data, it improved Sequencing towards that data. By This method both accelerates reads and ensures that an entire DNA pool need not be sequenced. They perform some lab experiments to check feasibility of their system in DNA. They performed various random access to read back only selected values. They further investigate in their design (method) using various computer stimulation to understand the error correction characteristic of different encoding schemes, access their overheads and make future based on technology trends. Growth in sequencing productivity eclipses even more’s law.

Effort done in this field

They believed that the DNA storage system is last and golden key of a deep as well as large storage problems.

Figure 1

Steps involved in dna storage

A DNA storage system consists of a DNA synthesizer that encodes the data into DNA Pool which is store into small compartment, and a DNA sequencer instrument that reads DNA sequences and convert them back into digital data (Figure 1). The basic unit of DNA storage has DNA strand that has roughly 100-200 nucleotide long which is capable of storing 50-100 bits total. The DNA strands was stored into “DNA pools” that have stochastic spatial organization and like hard disk and cd it does not permit structure addressing. Therefore, it is necessary to add the address itself into data stored in a strand.8, 9

Figure 2

Overview of a DNA storage system operation as a key value store.

Above figure is shown flowchart for input and output process of DNA storage in more detail. The write (input) process (Figure 2) needs key and value to store input. Key is useful for addressing and to determine the pool in the DNA library where the resulting strands storedand themultiple strands generated by the value. The primer target sequences, to produce final DNA sequence to be synthesized. The resulting DNA molecule is store into DNA library for future.The read (output) process, they needed to key. It is useful into obtained of the PCR primer sequence which was identify the key associated with DNA pool. The sample and PCR primer were sent to the PCR thermocycler, by use of thermocycler they amplified the desired strands. The resulting pool were further processin the DNA sequences, which produced the digital data readout. In the reading process there are some losses of sample of DNA from the pool So, it reduced quantity of DNA But DNA was easy to replicated, and so the pools can easily be restoring after read operations if needed. Whole DNA pool can be re synthesize after reading process.

Continues efforts which is done by various scientist

J. bronholt et al.:

The nucleotide is main base of the Data storage system. Its organic molecule consisting of one base (A, C, G, T) and Sugar Phosphate. This storage system mainly based on these 4 bases.as the results of some famous scientist new approach to stored binary data in DNA. It was quite difficult but it possible by great effort of the scientist and nucleotide base pair. The Quaternary digit can then be mapped to DNA nucleotides by producing string of n/2 digit from binary bits. (ex. mapping 0,1,2,3, to A, C, G, T, respectively). For example, the binary string 011001 maps to the base 4 string 1201, and then to DNA sequence CGAC. However, the DNA sequences and synthesis are very complicated it arise manyerrors so, it requires a more careful encoding. Some error is eliminated or reduce by encoding binary data in base 3 instead of base 4.7 to avoid the repetitions of same nucleotide they maps ternary digit to DNA nucleotide. This encoding avoids homo polymers-repetitions of the same nucleotide that significantly increasing the chance of sequencing error.9

Figure 3

Encoding a stream of binary data as a stream of nucleotides. A Huffman code translates binary to ternary digits, and a rotating encoding translates ternary digits to nucleotide.

Because base 3 is not a multiple of base 2, mapping directly between the bases would be inefficient: 6 ternary digits (36 =729) can store 9 bits of data (29 =512), but waste 217 possible states. Instead, they used a Huffman code10 that maps each binary byte to either 5 or 6 ternary digits.

For example, the Huffman code maps the binary string 01100001

To the base-3 string 01112. The rotating nucleotide encoding Maps this string to the DNA sequence CTCTG. by the help of Huffman code, they map ASCII character to 5- digit string.

Data format

Another practical problem is that they don’t have any synthesis technology to synthesize small length of sequences of nucleotides. Data is existing in hundreds of bits therefore cannot be synthesize as single strand of DNA. DNA pool do not perform spatial isolation, and so they addedsome keys which is irrelevant to a single read operation.Isolation of interested molecule and exiting DNA storage technology sequence entire pool which increase significant cost and time. To solve these two problems, they organized data in DNA in similar fashion to Goldman et al.7

Figure 4

Overview of the DNA data encoding format. After translating to nucleotides, the stream is divided into stands. Each strand containing payload from the stream, together with addressing information to identify the strand and primer targets necessary for PCR and sequencing.

Above Figure 4 shown the segment of nucleotide is divided into the block, which they synthesize a separate strand, so they get large storage capacity. Connect those strands with the identifying primers allows the read process to isolate the main interest of data molecule and so perform random Access. They add these different keys into our DNA sequence:

Encoding system for storage

In previous section study about organization of DNA storage system and how they store information by blocking system. They store a data in to DNA by broken strands of nucleotide sequences. It relies on the robustness of DNA for durability because each bit of data is encoded in exactly one location in the output DNA. Some Early work done by scientist they used simpler encodings technique For example, Bancroft et al.2, 5 translate text to DNA by means of a simple ternary encoding: each of the 26 English characters and a space character maps to a sequence of three nucleotides drawn from A, C, and T (so exactly 33 = 27 characters can be represented). They successfully recovered a message of 106 characters, but this encoding suffers substantial overheads and poor reliability for longer messages.

Goldman encoding

Let’s focus on an existing encoding proposed by Goldman et al.7 shown in Figure. This encoding is divided DNA nucleotide into overlapping segment to provide four-fold redundancy for each segment. This encoding provides high reliability. The Goldman used this encoding to successfully recover a 739 Kb message. He uses this encoding as a baseline because that time it is most popular DNA techniquein addition, it offers a tunable level of redundancy, by reducing the width of the segments and therefore repeating them more often in strands of the same length.

The experiment

Goldman and his team are work done on high capacity, low maintenance storage of digital information in synthesized DNA. They encoded computer files sizes of 739 KB of hard disk storage by help of Shannon information11 of 5.2*106 bits into a DNA code and synthesized these DNA sequenced it and redeveloped the original files with 100% accuracy.8 A series of experiments and their results proves DNA storage to be a realistic technology for large scale digital archiving that may already be cost effective for low access.

They understand and study the other DNA storage approaches problem. They developed an in-vitro approaches that represents the information is stared as a long DNA molecule and encodes this using shorter DNA fragment as same as church et al.6 They selected computer files and then encoded this into DNA. The five files comprised all 154 of Shakespeare’s sonnets (ASCII text), a classic scientific paper.12 (pdf format), a medium resolution color photograph of the European bioinformatics institutes (JPEG 2000 format), a 265 exert from martin Luther king’s 1963 “I have a Dream” speech (MP3 format). They used Huffman code to study to convert bytes to base 3- digits (ASCII text), and that produce a total of 757,051 bytes (Shannon information (11, 13) 5.2*106 bits). These five files were represented by a total of 153,335 strings of DNA, each string is comprising 117-nt.13

Figure 5

Image file of European bioinformatics institutes.

Figure 6

Digital information encoding in DNA

Digital information (a, in blue), here binary digits holding the ASCII codes for part of Shakespeare’s sonnet 18, was converted to base-3 (b, red) using a Huffman code that replaces each byte with five or six base-3 digits (trits).14 This in turn was converted in silico to our DNA code (c, green) by replacement of each digit with one of the three nucleotides different from the previous one used, so, there is ensure that no homopolymers were generate at this basis formation of large number of overlapping segments of length of 100 with 75 base pair is occur, creating fourfold redundancy (d, green and, with alternate segments reverse complemented for added data security, violet). Indexing DNA codes were added (yellow), also encoded as non-repeating DNA nucleotides. An additional advantage of their encoding scheme that the fragment length is perfect and uniform and absence of Homopolymers. So, obviously the synthesized DNA does not have a natural (biological) origin and the presence of aimful design encoded information.15 They designed DNA strings using an updated version of Agilent technologies OLS (Oligo Library Synthesis) process.16 They created a large number ˷2.5*106 of copies of each DNA string, with low error (1 error per 500 bases). Then they supply a lyophilized to synthesized DNA for excellent long-term preservation characteristics17, 18 and then this synthesized DNA was shipped (at ambient temperature, specified packaging) from the USA to Germany via the U.K and then they performed resuspension, amplification and purification a sample of resulting library product. Then it was sequenced in paired end mode on an Illumina HIseq 2000 and it was transferred to multiple aliquots and re-lyophilized for long term storage. the full length (117-nt) DNA strings were reconstructed in silico from the read pairs, with those containing uncertainties due to synthesis or sequencing error being discard by using reverse procedure of encoding. This discard string has information is recovered with more sophisticated decoding. So, they prove that DNA storage a potential as a practical solution to the digital archiving problem and may become a cost-effective solution for recovery assessed archives.

Figure 7

Illumina HIseq 2000

Figure 8

PCR thermo cycler

XOR encoding

While Goldman encoding provide high reliability. It suffers significant overhead: each block in the input string is repeated four types. They propose a simple new encoding that provides similar levels of redundancy to prior work, but with reduced overhead.Encoding shown in figure They tookA⊕B as payload A and B had two strands Which produce new payload. the high bit of the address is used to indicate whether a strand is an original payload or an exclusive or strand. These provided redundancy. Any of this payload is sufficient for recover third.

Figure 9

Our proposed encoding incorporates redundancy by taking the exclusive or of two payloads to form a third. Recovering ant two of the three stands is sufficient to recover the third.

James B et al. experiments:

They perform the experiment on random access capability of DNA storage. they encoded four image files using two different encoding method.

They performed experiment that they took various in files size from 5Kb to 84 Kb. They synthesize and sequencing these files and resulting DNA to recover the files. They used four images’ files for input to the DNA based storage system for each image file x. jpg, they generated DNA sequence related to the output of images (x. jpg…). They perform the experiments using two methods.

Figure 10

Three images files we synthesized and sequenced for our experiments.

They performed this experiment on four images, they used Goldman encoding method for three images and other one is encoded by XOR encoding system. (the Sydney. jpg image). Combine the 8 practical produce 45652 sequences of length 120 nucleotides Represents 151 Kb of datato demonstrate that DNA based storage system allows to effective random access. They synthesized sequence were prepared for sequencing by amplification via the PCR method. The product was sequenced using on Illumina Mises platform. The selected get operation total 16,994 sequences and 42 Kb produced 20.8 M reads of sequences in the pool.8 They inspected the result and observed no. of reads of sequence that were not selected. So, random access was effective in amplifying only target files. They successfully recovered all four files from sequenced DNA. They conclude that the sequencing depth is reduced so it will give batter results.8

Church et al.:

They developed strategy to encoded arbitatory digital information by using a novel encoding scheme that uses next generation DNA synthesis and sequencing technology. They converted HTML coded draft of a book that included 53,426 words of jpg images and one java script program into 5.27 mega bite bitstream.4 Then they encoded these bits onto 54,898.14 159-nt oligonucleotides, each encoding 96-bit data block (96-nt), A 19-bit address specifying the location of the data block in the bits stream (19-nt) and flanking of 22-nt common sequences for amplification and sequencing. This DNA library pool is synthesized by ink-jet printer, highly fidelity DNA microchips.14, 16 To read this encoded DNA it is necessary to be amplified the DNA library by limited cycle of PCR in thermo cycler and then sequenced on single lane of an Illumina Hi seq. Then they joined overlapping paired end 100-nt. Reads to reduce the effect of sequencing error.19

Figure 11

Comparison to other technologies. We plotted information density (log10 of bits/mm3) versus Current scalability as measured by the log10 of bits encoded in the report or commercial unit.

Then only expected 115-nt length and perfect barcode sequences the generated consequence at each base of each data block at an Avg of ~3000-fold coverage. (Fig). All data blocks were recovered with a total of 10 bits error out of 5.27 million (fig). Their method has at least 5 adv. over past DNA storage approaches. They encoded one bit per base instead of two (A or C for 0, G or T for 1). So, they can encode message many ways in order to avoid sequences that are difficult to read or write. By divided the bitstream into address data blocks, they eliminate the need for long DNA constructs. That are difficult to assemble at this scale. They synthesized, store and sequenced many copies of each individual oligo. They use purely in-vitro approaches that avoids cloning and stability issue of in-vivo approaches.

So, by this experiment they concluded next generation technologies in both DNA synthesize and sequencing to allow for encoding and decoding of large amount of information for 1, 00,000-fold less cost than first generation encoding.

Leon anavy et al.:

Oligonucleotide multiplicity, which is an important inherent property of current DNA synthesis and sequencing technologies is not exploited by the aforementioned work. They introduced c DNA letters that constructs and utilize this multiplicity and so, they able to increase the information capacity per synthesized portion. a composite DNA letter is a representation of a position in a sequence that constitutes a mixture of all four standard DNA nucleotides in a specified predetermine ratio. They describe that a composite DNA letters from the basis to a DNA synthesis approach that trades sequence multiplicity for increased the complexity of synthesized DNA effectively and it has higher data capacity per synthesized position. In the early days of DNA sequencings by hybridization, degenerate and semi-degenerate bases were proposed as wildcards for increasing the fidelity of the system.20, 21, 22 Next generation DNA sequence have higher quality and capacity when using degenerate base addition together with error correction approaches.23 They demonstrated practical on implementation of a complete large-scale composite DNA storage system by demonstration commercially available DNA synthesis and sequencesing techniques. Their method is superior to the previous method. They improved capacity of system implements an error correction scheme that combines an adaption of the previously repeated fountain code.24 They used composite DNA coding system to repeat the original DNA fountain experiment and increased 24% capacity per synthesized position. They stared a composite file contain an HTML version of the bible in both Hebrew and English taken from the mamre institute.25

Figure 12

Encodinga binary message using standard and composite DNA.

The experiment

They used this equation to calculate capacity for storage information:

Capacity=Length of binary message i after huffman codingLength of composite message i after huffman coding

They analysed all performance of function of process to better evaluate challenges and improvements of composite DNA based data storage. They performed a large-scale molecular implementation of a six-letter composite alphabet storage system. This encoded DNA is successfully retrieving the same 2.12 MB data file from erlich et al.24 That DNA pool consisted of 58,000 six letter composite oligo of length 152-nt, compared to 72,000 oligoes of same length required using standard DNA and then they increased 24% information carrying capacity per synthesized position and they make decoding pipeline that is allowing the correction systematics synthesis biases. We understand this pipeline below.

A compressed input file is being processed by the fountain code to produce binary droplets. A composite DNA encoding flow is then applied on each droplet consisting of the following steps (See Online Methods for details): (is) the binary message is translated into a composite DNA sequence. The seed sequence is translated to standard DNA sequence, which will serve as a barcode for the decoding process. The payload is translated to a six-letter composite DNA alphabet (Σ6) in 5-bit chunks. (ii) Error correction nucleotides are added to the DNA sequence by using a systematic Reed-Solomon (RS) encoding. The barcode is encoded using RS over 𝐺𝐹(13) and the payload is padded and encoded using RS over 𝐺𝐹. (iii) Each encoded message is then filtered to verify that the RS redundancy letters are all from Σ6. (iv) Experiment identifier and amplification template sequences are appended to each valid sequence. They also examined the minimal sequencing depth required to decode the message correctly for each one of the four-composite alphabet for higher resolution required deeper sequencing. They proved the concept of composite DNA, the properties of current DNA synthesize and sequencing process, to potentially attain higher density DNA based storage system. They proved and improved and implement in other approaches to increase capacity and fidelity of DNA based storage system, such as orthogonal base pair system,26 efficient coding13, 24, 27, 28 andrandom-access approaches13, 29, 30, 31, 32 incorporating composite DNA based storage system will require further investment in future.

Figure 13

Encoding pipeline of a large-scale composite DNA based data storage

M. blawat et al.:

M. blawat and his team also worked on a storage capacity boost and scheme for error correction method. They reported that strong capacity boost strength of storing digital data in synthetic DNA. They also developed an efficient and robust forward error correcting scheme adapted. To the DNA channel they used designed DNA channel model on data from a proof of concept conducted 2012 by a team from the Harvard medical school.6 They introduce their own method or scheme which is eliminate the all type of error of today’s DNA synthesis, amplification and sequencing process ex. Insertion, deletion and swap error by use of their method or scheme.33 They able to store and retrieve error free 22 Mbyte of digital data in synthetic DNA recently.(34) They also proves that the practically uses of synthetic DNA as long -term Digital data storage system. They analysis of the experiment data of church and his team gathered and produced a new designed forward error correction (FEC) scheme.6 Ex. Insertion and deletion and swap. They observed one type of Swap error occurred in an oligo, if a nucleotide had been replaced with incorrect one, at that time oligo length stays unchanged. so, an insertion or deletion error occurred in oligo, if their an addition nucleotide has been inserted or removed, the predominant error type the affect oligo corresponds lengthened or shorted. In the experiment data of church and his team. They found that the swap error rate lies between 6.0*10-4& 1.4*10-3, while insertion and deletation error rates are 1.0*10-3, and 5.0*10-3respectively (Blawat). They also fined some sequencing, which is not fined in read sequencing is called “missing oligoes”. And they also prove that the DNA storage system is not a memoryless data channel.

Figure 14

Experimentassumption of Blawat M. et. al.

They synthesized 900 000 230 nt oligonucleotides on Agilent’s oligo library synthesis (OLS) microarray platform, divided into four libraries with 225000 oligonucleotides in 100 µL TE. Illumina specific sequencing adaptors were introduced into the synthesized OLC pool in a two-stage serial PCR amplification using the syBR FAST MASTER mix. Reaction was performed using the following protocol on an Eppendorf master cycler realplex 4 real time PCR machine by monitoring the syBR green channel signal.33 Each reaction was harvested after and cycle of amplification to avoid PCR bias in the resulting library the resulting PCR products after each stage were purified using agencourt amppureXP beads according to manufacturer’s instruction.

Figure 15

Illumina HIseq. 2500 next generation sequencer.

Figure 16

ThesyBR FAST MASTER Mix.

They sequenced the amplified library by loading 1mL of 16mL library on 2 LANs of a rapid sequence 300 cycle SBS kit an Illumina HIseq. 2500 next generation sequencer. They obtained 144,475,005 paired reads with 83.78 % of the reads scoring >=Q30. They are successfully work done on storage capacity boost and sequencing of DNA.33

We discussed about various DNA data storage approaches which has important role into future of digital data storage into DNA they work at all learn problem and get their solution and give some more efficient, and secure technique is provide to us for our better future.

Application of digital data storage DNA fountain