What are the challenges of creating synthetic human genomes?

[1oldman2]

I am interested in hearing more on this, Pro's Con's etc. Any thoughts or info would be greatly appreciated.
http://www.nytimes.com/2016/05/14/science/synthetic-human-genome.html?_r=0

 

[Ygggdrasil]

We are very far away from this goal. First, the technologies for DNA synthesis and assembly are not ready to synthesize the 3 billion base pairs of the human genome. Scientists have synthesized a bacterial genome (0.0005 billion base pairs) and a yeast chromosome (0.0002 billion base pairs). The technology might be ready in a decade, but it is possible that other methods of genome engineering and editing (such as https://www.physicsforums.com/insights/dont-fear-https://www.physicsforums.com/insights/dont-fear-crispr-new-gene-editing-technologies-wont-lead-designer-babies/-new-gene-editing-technologies-wont-lead-designer-babies/) would enable similar experiments within a decade as well. A better starting point would be to finish creating the synthetic yeast genome, which would provide a nice platform for doing many of the studies one might want to do on a synthetic human genome. However, there are things one could learn from studying synthetic human genomes, that one could not learn from studying synthetic yeast genomes (though synthetic yeast genomes would definitely have more potential for applications in biotechnology).

Second, on the biology side, we also lack the knowledge to rationally design genomes. Earlier this year, researchers published a paper on synthesizing a bacterial genome in which they tried to design the bacterium to have as few genes as possible. Attempts to rationally design such a bacterium failed, and the team had to resort to very tedious trial-and-error approach. So, even if we were to have the capability to fully synthesize a human genome, we would not be able to engineer super soldiers as the NY Times article suggests. Similarly, genetics are not destiny and environmental factors play a huge role in an individual's development. Synthesizing Einstein's genome would likely not produce another genius.

Scientifically, there are some interesting things that we could learn from such a technology, but whether the synthesis of human genomes enables any practical applications is very unclear.

 

[rootone]

Typically the function of a gene is to act as instructions for building a protein.
The protein then is used to build some aspect of the physical organism.
The detailed structure of the proteins differ between individuals (which is why individual examples of the same species are not identical)
Sometimes a variation of a protein doesn't work very well and the organism may not be able to survive, or will suffer from a disability.
At the present time we don't know a lot about the function of some proteins, so therefore attempts to make new variations do not lead to a predictable result.
There has been some success though with replacing genes which are known to be dysfunctional with another sequence which is known to be OK.

 

[Ygggdrasil]

IMHO, the benefit synthetic human (or mammalian) genomes for research purposes is not in studying protein function, but in studying non-coding DNA.

>98% of the human genome does not code for protein, and we don't know the function of many of these sequences. Some of these sequences have been called "junk DNA," but scientists debate about how much of the human genome is junk DNA. To what extent can some of these junk sequences be removed while still maintaining the ability to encode a fully functioning organism? Attempts to synthesize "minimal" mammalian genomes could help answer this question and clarify the function of "junk DNA."

Some of the non-coding sequences control which protein-coding genes are turned on and off, but many of these regulatory sequences are far away from the genes they regulate. A big area of research is in figuring out the three-dimensional architecture of DNA within the nucleus and whether the physical organization of DNA plays a role in controlling gene expression (for example, by folding the DNA such that the regulatory sequences touch the genes they regulate but not genes that should be maintained off). The ability to synthesize new chromosomes could allow scientists to directly test ideas about how non-coding sequences influence genome architecture and gene expression.

Enabling experiments to address the above questions (as well as many others) would be the main scientific benefit of such research. I still am skeptical of any practical application for synthesizing human genomes other than helping to advance technologies that might help in other synthetic biology efforts (e.g. in engineering bacteria or yeast for applications in the biotechnology industry).

 

[1oldman2]

Thanks for the replies/opinions, From the layman's point of view its very difficult to get an idea of where this technology is based on a news article such as I have cited here. The concepts the article mentions are very interesting but I felt bringing it up in a thread here would be a good reality check. This is the best possible place I can seek an opinion that is reliable. Did I mention I love this site? Thanks again.

 

[Ygggdrasil]

Today, the HGP-write team published an article online in the journal Science discussing some of their plans: http://science.sciencemag.org/content/early/2016/06/01/science.aaf6850.full

A series of pilot projects making use of very long DNA sequences that are nonetheless short of a full genome are anticipated: (i) synthesizing “full” gene loci with accompanying noncoding DNA to help explain still-enigmatic roles of noncoding DNA variants in regulating gene expression, and leading to more comprehensive models for the role of noncoding genetic variation in common human diseases and traits; (ii) constructing specific chromosomes—e.g., chromosome 21—or complex cancer genotypes to more comprehensively model human disease; (iii) producing specialized chromosomes encoding one or several pathways—e.g., all genes needed to make a prototrophic human cell, or pathways to transform the pig genome to make it more amenable as source for human organ transplantation; (iv) a potential transformation of gene therapy, with freedom to deliver many genes and control circuits to improve safety and efficacy, provided delivery challenges can be met. Indeed, many substantial and useful innovations may be realized in such “stepping stone” projects that are short of whole-genome re-engineering but require substantial improvement in synthesis capacity of Mb- to Gb-sized DNA. [...]

Additional pilot projects being considered include (v) using induced pluripotent stem cells (16) to construct an “ultrasafe” human cell line via comprehensive recoding of protein-coding regions, and deletion of corresponding genome features to increase safety of such a cell line (see Box 1); and (vi) developing a homozygous reference genome bearing the most common pan-human allele (or allele ancestral to a given human population) at each position to develop cells powered by “baseline” human genomes. Comparison to this baseline will aid in dissecting complex phenotypes such as disease susceptibility.

Boeke, Church, Hessel, Kelley et al. 2016. The Genome Project–Write. Science. Published online 02 Jun 2016. doi:10.1126/science.aaf6850

They've also launched a webpage for their project: http://engineeringbiologycenter.org/

 

[1oldman2]

http://science.sciencemag.org/content/early/2016/06/01/science.aaf6850.full

This is great, I'm still reading and absorbing it.

http://engineeringbiologycenter.org/

 

[1oldman2]

http://www.genengnews.com/gen-news-highlights/let-s-synthesize-the-human-genome-says-hgp-write/81252790/
This seems to be getting a lot of attention (as well it should) in the Biomed community. I expect a lot of "doomsaying" in certain circles but I think this will be a great step in the right direction for the cure and even elimination of many diseases and conditions currently prevalent. Once again I need to point out my thoughts here are based on a "laymans" point of view, I'm sure that there are many aspects of the technology I will be hearing about soon, Thanks.

This one I'll be considering for a while but thought I'd post the link now.
http://www.scientificamerican.com/article/plan-to-synthesize-human-genome-triggers-a-mixed-response/

 

[Ygggdrasil]

In my opinion, the rationale for synthesizing a human genome is similar to the rationale for going to the moon. The goal itself is big and flashy, and sure to draw a lot of attention and publicity, but the scientific payoff is not in the goal itself, but in the technologies we develop in the process of achieving the goal. I do not expect synthesizing human genomes to cure or eliminate any diseases in the next few decades. However, some of the spin off technologies and the results of the scientific research it enables will aid in our understanding of the fundamentals of biology and may bring us closer to learning how to cure or eliminate some diseases.

 

[1oldman2]

In my opinion, the rationale for synthesizing a human genome is similar to the rationale for going to the moon. The goal itself is big and flashy, and sure to draw a lot of attention and publicity, but the scientific payoff is not in the goal itself, but in the technologies we develop in the process of achieving the goal. I do not expect synthesizing human genomes to cure or eliminate any diseases in the next few decades. However, some of the spin off technologies and the results of the scientific research it enables will aid in our understanding of the fundamentals of biology and may bring us closer to learning how to cure or eliminate some diseases.

This is a much more accurate than my statement, thanks for the perspective.

 

[mfb]

Where do you see the main challenges in that project? Is it just scaling up by a factor of 10000, or are there completely new challenges putting longer and longer gene sequences together?

 

[1oldman2]

Where do you see the main challenges in that project? Is it just scaling up by a factor of 10000, or are there completely new challenges putting longer and longer gene sequences together?

I have considerable reading ahead of me to give an answer worth posting.

 

[Chronos]

Some big questions remain to be solved. We do not yet grasp the synergy between genes. Apparently unrelated genes may team up to produce subtle but vital biological functions. Also, even some 'bad' genes behave unpredictably. The difference between sickle cell anemia and malaria resistance is one copy of the same gene. One faulty copy of the CFTR gene is harmless and possibly even beneficial, two copies produce cystic fibrosis. Then there is the matter of timing. Certain genes turn on and off at different developmental stages of an organism - from inception through adulthood. Turning the wrong gene on or off at the wrong time can have dire consequences. Until we better understand the interplay and intraplay between genes, their manipulation is a perilous exercise.

 

[mfb]

I have considerable reading ahead of me to give an answer worth posting.

My question was mainly directed at @Ygggdrasil. Your post was not there yet when I opened the thread.

Until we better understand the interplay and intraplay between genes, their manipulation is a perilous exercise.

Their manipulation is one way to advance our understanding.

 

[Ygggdrasil]

Where do you see the main challenges in that project? Is it just scaling up by a factor of 10000, or are there completely new challenges putting longer and longer gene sequences together?

Scaling up the capacity to synthesize and work with DNA is certainly a major challenge, but it is not the only challenge. If the goal is simply to copy an existing human genome, we will have to find ways to introduce the newly synthesized chromosomes into an enucleated cell and get them "booted up" correctly. This is not a trivial task given that eukaryotic gene expression is regulated not just by DNA sequence but by the protein and small RNA factors that allow the DNA to be read correctly. There are some fundamental questions, for example, as to the importance of DNA methylation, chromatin structure, and histone modifications, that need to be answered in getting the synthetic genome to function. There is reason to think a small number of factors could enable the genome to boot up (e.g. as in the case of the Yamanaka cell reprogramming experiment), but the situation would be very different starting from a completely synthetic, unmodified and unpackaged genome.

If the goal is to begin designing new human genomes with new functions, this goal is much more than a decade away. For example, it took Venter six years to go from bacterial genome synthesis to a "minimal" bacterial genome, and even then they couldn't rationally design the minimal bacterial genome. In the end, Venter's efforts were hampered by—and revealed in a very stark way—the fact that we don't understand all of the protein-coding sequences required for life. In eukaryotes, the problem is even more complicated. Not only do we have more genes, but there are many more regulatory non-coding elements in the human genome, and identifying and decoding their function is much more difficult than in the case of protein-coding sequences. Hopefully, the HGP-write effort will spur on progress in this field by developing new tools to understand the function of non-coding genetic elements, but I don't see these efforts translating to practical applications of human genome synthesis anytime soon.

 

[1oldman2]

My question was mainly directed at @Ygggdrasil. Your post was not there yet when I opened the thread.

After reading post #11 I assumed you were likely addressing Y, however I wasn't positive if that was the case or were you asking me a sort of "test question" to see if I was doing my reading. I choose to "error on the side of caution" and post some sort of response rather than appear to be dodging a valid question.
I just wanted to explain my post.
At this point the subject is way above me, I'll resort to following posts and see where this goes. Thanks much.

 

[CynicusRex]

I thought we were much closer to positive eugenetics than described here. Nice read.

>98% of the human genome does not code for protein, and we don't know the function of many of these sequences. Some of these sequences have been called "junk DNA," but scientists debate about how much of the human genome is junk DNA. To what extent can some of these junk sequences be removed while still maintaining the ability to encode a fully functioning organism? Attempts to synthesize "minimal" mammalian genomes could help answer this question and clarify the function of "junk DNA."

If anyone is interested, my friend is doing a PhD on ribosome profiling data. The excerpt below comes from his website: http://sorfs.org/

Introduction
Small open reading frames (sORFs) can be defined as open reading frames smaller than or equal to 300 nucleotides (100 amino acids). These “sORFs”, while inherent to all genomes, are historically ignored in gene annotation studies, stating that these lack any coding potential. Exclusion of these sORFs has emerged as a side effect during the development of different (gene prediction) tools in the field of bioinformatics/genomics/proteomics trying to reduce noise, imposed by technological limitations However, recent scientific breakthroughs discovered coding potential of several sORFs with clinical significance, indicating their importance. 1, 2, 4 . In particular, the advent of ribosome profiling 5 (RIBO-seq), a next generation deep sequencing technique, providing a genome-wide snapshot of the translating machinery in a cell, provided evidence of translation in sORFs. The value and importance of sORFs is becoming widely recognized 6, 7 furthermore ribosome profiling data is becoming more abundant. The creation of a public repository for sORFs, providing information resulting from various tools and metrics, seems a necessity in aiding functional research in the micropeptide field.

What does the database hold:
With this in mind, we like to introduce sORF.org, a public repository for sORFs. The main purpose is to allow researchers to examine individual sORFs or to perform searches based on several criteria for further large-scale studies. Different data sources, both experimental and in silico(based on various bioinformatics tools), are collected. sORF.org currently holds 333970 sORFs across three different species (human, mouse and fruit fly), derived from multiple RIBO-seq experiments and is expanding as more data becomes available. Available datasets can be inspectedHERE.

 

[dipole]

Scaling up the capacity to synthesize and work with DNA is certainly a major challenge, but it is not the only challenge. If the goal is simply to copy an existing human genome, we will have to find ways to introduce the newly synthesized chromosomes into an enucleated cell and get them "booted up" correctly. This is not a trivial task given that eukaryotic gene expression is regulated not just by DNA sequence but by the protein and small RNA factors that allow the DNA to be read correctly. There are some fundamental questions, for example, as to the importance of DNA methylation, chromatin structure, and histone modifications, that need to be answered in getting the synthetic genome to function. There is reason to think a small number of factors could enable the genome to boot up (e.g. as in the case of the Yamanaka cell reprogramming experiment), but the situation would be very different starting from a completely synthetic, unmodified and unpackaged genome.

If the goal is to begin designing new human genomes with new functions, this goal is much more than a decade away. For example, it took Venter six years to go from bacterial genome synthesis to a "minimal" bacterial genome, and even then they couldn't rationally design the minimal bacterial genome. In the end, Venter's efforts were hampered by—and revealed in a very stark way—the fact that we don't understand all of the protein-coding sequences required for life. In eukaryotes, the problem is even more complicated. Not only do we have more genes, but there are many more regulatory non-coding elements in the human genome, and identifying and decoding their function is much more difficult than in the case of protein-coding sequences. Hopefully, the HGP-write effort will spur on progress in this field by developing new tools to understand the function of non-coding genetic elements, but I don't see these efforts translating to practical applications of human genome synthesis anytime soon.

Well, using the moon analogy, just like the Apollo missions were preceded by Mercury and Gemini - with smaller goals of putting a man in orbit before we put one on the moon, I'm sure the Synthetic Human Genome project would start with a smaller step first. Perhaps synthesizing a prokaryote genome, or a simpler eukaryote (yeast) genome.

I fully support this idea. If the LHC was worth sinking billions of dollars into, then this certainly is worth pursuing.

 

[Ygggdrasil]

Well, using the moon analogy, just like the Apollo missions were preceded by Mercury and Gemini - with smaller goals of putting a man in orbit before we put one on the moon, I'm sure the Synthetic Human Genome project would start with a smaller step first. Perhaps synthesizing a prokaryote genome, or a simpler eukaryote (yeast) genome.

I fully support this idea. If the LHC was worth sinking billions of dollars into, then this certainly is worth pursuing.

A full prokaryotic genome has already been synthesized six years ago: https://www.physicsforums.com/threads/bacterial-cell-with-a-chemically-synthesized-genome.404603/

Scientists reported the synthesis of a yeast chromosome two years ago: http://www.nature.com/news/first-synthetic-yeast-chromosome-revealed-1.14941

The synthetic yeast genome will likely be done in a year or two: http://syntheticyeast.org/sc2-0/

While some of the sub-goals of the synthetic human genome project are definitely worth funding (I am particularly fond of their Aim I: synthesizing “full” gene loci with accompanying noncoding DNA to help explain still-enigmatic roles of noncoding DNA variants in regulating gene expression, and leading to more comprehensive models for the role of noncoding genetic variation in common human diseases and traits), I am not sure a "big science" approach is the correct approach to take. Science would probably move forward better by giving out many smaller grants to individual investigators rather than trying to bring everyone into one centralized endeavor. Big science works well when the means of going forward are clear. There are still many fundamental challenges that need to be solved, so it's better to have many labs exploring a bunch of different ideas on how to go forward than centralizing everything.

For example, one can look at the example of the human genome project. The NIH centralized the whole effort and was making very slow progress for many years. It wasn't until Craig Venter with Celera Genomics jumped in as outsiders with new ideas that things began progressing faster.

 

[Fooality]

To what extent can some of these junk sequences be removed while still maintaining the ability to encode a fully functioning organism? Attempts to synthesize "minimal" mammalian genomes could help answer this question and clarify the function of "junk DNA."

I'm no biology guru. Is it possible that there is "junk" DNA which serves no purpose in creating a functional individual, but has a purpose in a larger evolutionary sense? For instance if modern cows started hanging out in water for millions of years until evolving into whale like aquatic creatures, is it possible that dormant junk DNA could be activated, to give them some benefits for aquatic life evolved in the distant past?

 

[Ygggdrasil]

I'm no biology guru. Is it possible that there is "junk" DNA which serves no purpose in creating a functional individual, but has a purpose in a larger evolutionary sense? For instance if modern cows started hanging out in water for millions of years until evolving into whale like aquatic creatures, is it possible that dormant junk DNA could be activated, to give them some benefits for aquatic life evolved in the distant past?

Although I'm not aware of any cases of dormant genes being reactivated in evolution (there probably are cases, though), there is some evidence that certain classes of "junk DNA" could aid in evolution. For example, transposable elements (TEs), also known as jumping genes or selfish genetic elements, have been proposed to do this:

The elements contain in their sequence all the instructions needed to cut themselves out of their host DNA and splice themselves into another spot. But they are not always benign 'junk' DNA — they can insert into genes or gene regulatory elements, potentially disrupting the gene's function, and they can trigger chromosome rearrangements. So, even though most copies are selectively neutral and not in themselves damaging, they have long been considered as predominantly harmful to their hosts, as they can contribute to the appearance of mutations, some of which can result in disease.

But TEs do not always have adverse effects, and their mutational activities contribute to the genetic diversity of the organism. Indeed, some TEs have been domesticated by their host genome, acting as genes or gene regulatory elements, and as a result constitute a source of genetic innovation for the organismhttp://www.nature.com/nature/journal/v443/n7111/full/443521a.html#B1http://www.nature.com/nature/journal/v443/n7111/full/443521a.html#B2. Progress in understanding how these elements are regulated is bringing an appreciation of how an individual's environment can affect the expression of their genetic complement to produce their own particular characteristics (phenotypes), such as physical appearance, behaviour, susceptibility to disease and even neuronal function.

http://www.nature.com/nature/journal/v443/n7111/full/443521a.html

 

[Fooality]

Wow, mind blown reading about that. Except now I have 100 times more questions than when I started, and I'm not sure anyone has answers.

Transposable elements (both active and inactive) occupy approximately half the human genome ... We now realize that some transposable elements are also viruses, ... Indeed, some viruses may be derived from natural transposable elements and vice versa. Since viruses move between individuals, at least some transposable elements can move between genomes (between individuals) as well as within an individual’s genome... It is not even clear whether transposable elements should be considered an integral part of a species’ genome, or if they are successful parasites.

http://biowiki.ucdavis.edu/Core/Gen...9._Transposition_of_DNA/Transposable_Elements

This makes me think of all the havoc Europeans brought on Native Americans through illness when they came, and it makes me think of HIV in Africa. If a population were ravished by HIV long enough, could it become a part of of them permanently? How much of that junk DNA is actually from viruses?

And looking into that I saw this:
http://www.popsci.com/science/article/2010-01/8-percent-human-dna-comes-virus-causes-schizophrenia

So these TE's, some of which are transmitted from animal to animal can potentially effect neuronal structure/cognition enough to create changes as dramatic as schizophrenia. That along with human race being 2 million years old, with civilization booming only in the last 3000 years, makes me ask this crazy question:

Could language, could civilization, be caused by a brain altering VIRUS!?

Ok, I'm going off into scifi space a little here, but that's the level of questions this stuff brings up to me. Thanks for sharing!

 

[write4u]

I don't think the problem lies in creating an exact synthetic copy of DNA. IMO, the problem lies in the duplication process. This may be of interest.
http://www.ted.com/talks/drew_berry_animations_of_unseeable_biology