Object oriented biological library

(UPDATED) It appears that MIT already has an open program and that OpenWetWare is non-profit. Link to MIT parts registry. That is awful , duplicate invention before I could even finish thinking about it. They are a bit ahead of me there. They still need a language to manage the information. Strange that they didn't think of that. It is too complex to deal with so many different parts without some structure to the tools. IMHO

This is interesting from wetware wiki.

For the sake of having a proper foundation for the biological language it seems that a library would be the best way to implement the sub structure. Since millions of random combinations could have some utility, the parts that were useful in a specific context would best be organized to be easily accessible in assembly. It would seem that if I were making an amino acid assembler that tRNA for each of the basic amino acids would be a must and the rRNA along with its associated proteins. It would seem that a power supply (ATP) would best be created separately as it is so common.

It wouldn't be necessary to have a mechanism for enclosure or export if it were created within a liquid computer, but certainly if it were to be exported to an asteroid it should be able to create all of its structure.

It wouldn't make sense to keep creating the mRNA or tRNA sequences each time that they were useful and a common mechanism for concatenation would be a primitive operator akin to "+".

The language need not be self documenting or even pretty, it is only the interface that should be simple to use. Since gene sequences are essentially strings and words in a language with four characters it would seem that string operators would be sensible and that a string have a marker that was unique like a handle that was its tuple element so that it could be operated upon. I might say &("Chromosome 1")+("cell wall attach sequence A")+("promoter glucose")+("glucose handler")+("stop sequence")+("cell wall attach sequence B"). I don't know whether chromosome 1 would be a good name to use. Since it derives from biological staining it doesn't really describe the intent of the design. It would seem that an object class or container name would be more relevant. Since it wouldn't actually be generated as a chromosome and couldn't be viewed as such it is a misnomer.

For the sake of convenience and using existing methods to accelerate the development it would make sense to start with a liquid computer mechanism and a basic set of operators that can be used commonly. DNA sequences that correspond to a controlling protein sequence as matched regulating pairs. It would seem that numerous matched pairs would be advisable and a choice of effectiveness could be used to determine the strength of regulation feed-thru or feed-back.

In conjunction it would seem wise to have a visual interface that would allow pick and place along with a simulator that mimicked the operation in a particular environment. As with cell free extracts, it is possible to model a process and then measure its response to adjust the algorithm for design. It would make sense to have access to all the variables possible in some sets. It might be advisable to consider a design as it operated in vacuum or at low temperatures or high acid concentrations. There would be no restriction to using specialized components manufactured without biological equivalent, however that would lead to problems if the goal was to extend the system at the speed of light within the available networks. Any complex device would have a cost at such a remote location that would seem to be prohibitive. I suppose if a standard computational control assembly were feasible then it would make no difference. It does seem that it would be better if no waste product that couldn't be recycled would be produced. That would lead to additional complexity of new designs.

In the case of the Internet and Earth it proceeds at the speed of light as all of the biological raw materials are available everywhere. As with a language like LISP it would be advisable to define valid primitives which can be combined to perform any task in a biological environment. I suppose that garbage collection or freeing unused memory would be similar to keeping raw material available. I think I would prefer an operator that does CONS and de-CONS in steps. The core design is simple in the fact that it has one method that adds character X to some sequence and so CONS. When decomposition is advisable, de-CONS or perhaps DECONS would be used on a specific container or object. There is no restriction that the code should be permanent like a BIOS or microcode, if the establishment is very different than the application it would not be necessary to store the program segments that would never again be used. It also serves to avoid complexity that would make further design difficult if each available site of regulation needed to be simulated to avoid interference.

There is ample documentation on antigen antibody pairing to establish self assembling object container systems. It is similar to monoclonal antibodies and simple methods are available for testing these. In a composite system the association of parts could be determined with great 3D precision. If there is some issue of certainty in the final structure it could perform something like a boundary scan or interaction with an apoptosis regulator that would restart the process from scratch if it failed final Q&A.

The design could be implemented in such a way that anyone could program the elements and design on the foundations of other previous capabilities. In the same way that I might start with a blender leg model and a blender arm model, I could combine them in any way I wanted so long as I understood the IDE to the system. So instead of creating virtual characters in a virtual world, it would be possible to create actual functioning organisms. I have already done designs for neural arrays and muscles as well as energy collectors. I don't see it as being any more complex than assigning a bone and actuator in blender. They are single purposed devices and if they are damaged they are not self repairing.

I don't see any great hope that it will be possible to reverse engineer the chaotic process of human genetic and biological structure, but rational design replacement seems possible. The simple fact that it incorporates siRNA in such vast quantities implies that it has so many kludges that the network circuit of the process would be a task completed at the "Restaurant at the End of the Universe".

I was going to make a joke about patenting the combination of nucleotides, but it seems that joke is just business as usual. Of course nobody is willing to respect my patent on the proton, but I am guessing if enough congressmen have a monetary interest in a genetics company they would be given exclusive rights to Drew Barrymore's DNA if they wanted to claim it.

0 comments:

Automated Intelligence

Automated Intelligence
Auftrag der unendlichen LOL katzen