--- Log opened Sat Mar 02 00:00:04 2019
01:46 -!- Gurkenglas [~Gurkengla@unaffiliated/gurkenglas] has joined ##hplusroadmap
01:55 -!- Urchin [~urchin@unaffiliated/urchin] has quit [Remote host closed the connection]
03:16 < fltrz> regarding question 1 (if there is exactly one homologous recombination crossover per chromosome per meiosis event), if true this may be related to the wikipedia statement " In many animals, the organization of the Hox genes in the chromosome is the same as the order of their expression along the anterior-posterior axis of the developing animal, and are thus said to display colinearity."
03:19 < fltrz> i.e. for each chromosome homologue, if the homologous crossover occurs at exactly one point along the chromosome, it indicates the sperm / egg cell will inherit (along the head-tail axis) a head-X from say grandfather and then X-tail from grandmother or vice versa)
03:20 < fltrz> where X is one of the recombination hotspots
03:24 < fltrz> I believe it is the one crossover per meiosis event resulting in consecutive contiguous regions of the 2 homologue chromosomes mechanism which has aligned these Hox genes, as this mechanism ensures a high level of compatibility, except for the crossover location
03:26 < fltrz> the potential genetic mismatch at the crossover location is remedied by having hotspots such that relatively stable "API" results, with stable division of genes between hotspot locations
03:32 < fltrz> I would like to compare the relative alignment of these Hox regions, insulating neighborhoods, recombination hotspots in a human genome, is there a recommended high quality freely available genome? craig venter's? some other?
03:50 < fltrz> if the head-tail and Hox gene position collinearity is imperfect, and there are 23 homologous chromosome, then the imperfect collinearity of each of those 23 chromosome types could correspond to a "coordinate axis" in the animal, i.e. groups of cell types in the same region or body part (perhaps not strictly linear, think skin, or blood vessel) may share similar hox protein expressions
03:51 < fltrz> so that theres about 23 coordinates per location type (and each location type would probably support multiple cell types, defined by stable cellular homeostasis points of Hox *and other genes*
03:55 < fltrz> i.e. cell pairs which are neighbors but in different body parts need to be compatible, and expression differences should only involve genes at most one hotspot separated on each chromosome, evolution would spontaneously avoid expression differences 2 or more hotspots away on any axis
03:58 -!- nsh [~lol@wikipedia/nsh] has quit [Killed (Sigyn (Spam is off topic on freenode.))]
03:58 -!- ensign [~ensign@2001:41d0:8:d711::1] has quit [K-Lined]
04:02 -!- ebowden [~ebowden@unaffiliated/ebowden] has quit [Ping timeout: 250 seconds]
04:06 -!- ebowden [~ebowden@unaffiliated/ebowden] has joined ##hplusroadmap
04:06 -!- ensign [~ensign@2001:41d0:8:d711::1] has joined ##hplusroadmap
04:13 -!- nsh [~lol@wikipedia/nsh] has joined ##hplusroadmap
04:24 < fltrz> this image seems to indicate that the chromosomal territories might be bulbous, i.e. even though the different territories are on the same side of the nuclear lamina the bulbs or protrusions provide some spatial isolation https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2829961/figure/A003889F1/
04:24 < fltrz> from https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2829961/
04:29 < fltrz> oh wow what they did was pure genius: laser UV radiation (257nm) in a small spot in the nucleus, if random all chromosomes would be affected, if chromosomal territories damage would concentrate to a low number of chromosomes
04:33 < fltrz> "These experiments provided the first compelling, although still indirect, evidence for the existence of chromosome territories." is this humility? thats pretty direct evidence in my book
05:03 -!- ebowden [~ebowden@unaffiliated/ebowden] has quit [Ping timeout: 250 seconds]
05:06 -!- ebowden [~ebowden@unaffiliated/ebowden] has joined ##hplusroadmap
05:31 < nsh> fltrz, you might get some clues here: http://ssgrr2002w.atspace.com/papers/181.pdf
05:31 < nsh> i'm sorry the maths and physics is quite involved but it might give you a taste of the complexity and nuance here
05:31 < fltrz> thanks
05:54 < fltrz> nsh, do you know if the optical readout of DNA base sequences as described on page 5 (or page 6 in a pdf reader) was successfully implemented? this seems very relevant to one of kanzure's DNA memory goals
05:57 < nsh> i'm doubtful anyone has done it but the technology is very much improved and if it was only conceivable when written it may be very possible today
05:58 < nsh> here are cites for a similar paper: https://scholar.google.com/scholar?client=ubuntu&hs=GWn&channel=fs&um=1&ie=UTF-8&lr&cites=15016685422235556814
05:58 < nsh> sorry that exact paper even
05:59 < nsh> this book might be worth a browse: http://booksdescr.org/item/index.php?md5=8ec8a4ac227f115714df650d350d0fe0
06:00 < nsh> the thing about the DNA molecule that people are taking a very long time to appreciate is that it must have some kind of klein bottle like non-orientable topology because there is a continuous relationship with the cell and even tissues and the entire organism and the DNA molecules within each cell
06:01 < nsh> this is what allows for the coherent orchestration of complex processes within and between cells or other functional structures
06:01 < nsh> it is reasonable to conjecture that this is mediated by a kind of holography
06:02 < nsh> holography as the creation of a higher dimensional projected coherent image from a lower dimensional medium is made possible by long range phase coherence across and about the medium
06:03 < nsh> so a thing to look for within the bioinformatics might be signatures of this long-range phase coherence
06:03 < nsh> or correlation lengths
06:03 < nsh> unfortunately as i was trying to get across yesterday this might not be captured by the more or less static and rigid backbone of the base pair helix
06:03 < nsh> because the coherence and phase correlations will be in the electron cloud spin states
06:04 < nsh> however just as a hologram fixes the potential to project coherence through the structure of the holographic plate medium, it might be the case that with sufficiently sophisticated physical modelling of the consequences of the DNA molecule and its folding dynamics
06:05 < nsh> you can demonstrate that just as a hologram maintains the coherence of the reference laser (or the spectral equivalent for rainbow holograms) and modulates it by the imaged 3d surface
06:05 < nsh> that the DNA molecule is capable of doing the same
06:06 < nsh> i'm just not sure we can do the physics computationally with sufficient fidelity to show this
06:06 < nsh> it might be easier just to do it via the crystallography in vivo/vitro
06:07 < nsh> but it might be necessary to more or less invent the field of quantum holographic molecular crystalography and even without the crystalisation which might interfere with the effects
06:07 < nsh> but who said all of this was going to be easy
06:08 < nsh> there's also the work of Matti Pitkanen but he's a bit... well, somewhere between vanilla incomprehensible and borderline kooky
06:08 < nsh> e.g. https://www.researchgate.net/publication/277182525_Model_for_the_Findings_about_Hologram_Generating_Properties_of_DNA
06:13 < nsh> [37] P. Gariaev et al (2000), The DNA-wave-biocomputer, CASYS’2000, Fourth International Confer-
06:13 < nsh> ence on Computing Anticipatory Systems, Liege, 2000. Abstract Book, Ed. M. Dubois.
06:14 < nsh> Gariaev's Russian-language site is available on the wayback machine: http://web.archive.org/web/20090429110930/http://www.wavegenetics.jino-net.ru/
06:14 < nsh> vaguely readable after machine translation
06:15 < nsh> although they maybe didn't save the zip archives :(
06:16 < nsh> i feel like we have a friend or two at the internet archive who might be able to tell us if they do save them behind the scenes
06:17 < fltrz> I understand the usual interpretation of holography and diffraction patterns, but I only see how it can be used to read out a structure, I'm a bit baffled by the idea that holography explains the functionality of DNA in vivo though
06:17 < nsh> ah, some of them are saved
06:17 < nsh> http://web.archive.org/web/*/http://www.wavegenetics.jino-net.ru/zip/*
06:17 < nsh> don't be surprised. it's not an idea that has any currency yet
06:18 < nsh> and unfortunately the few people who discuss it do not do so in a very or even remotely accessible manner and i can't especially do it justice in paraphrasal
06:18 < nsh> the easiest thing to do would probably be to create a dynamic holographically constrained system as a proof of concept
06:19 < nsh> which i'm working on
06:20 < nsh> the other approach would be to do a literature review of some recent concepts in neural networks and theories of human visual perception
06:20 < fltrz> consider a camera system, to understand how lenses work, and how a photographic plate can record the light intensity is one thing, but that does not tell us much about the subject matter in a picture: when we take pictures of automobiles say, we know that the light struck the automobile, reflected and was focused by the lens on the photographic film, but this knowledge about the camera does not tell us very much about automobiles
06:20 < nsh> there is a notion of information representation that is amenable in neural networks called "holographic reduced representations"
06:21 < fltrz> you can learn a lot about automobiles (brands, other vehicles on the road, timeframe, ...) by looking at such pictures, but the exact imaging or readout mechanism seems irrelevant to the subject matter
06:21 < nsh> so there is a way of viewing the dynamics of a neural network as a mutually interpenetrating set of holographs reproducing new holographs in the distributed representations of data contained by the network
06:22 < nsh> yes but we're talking about a car that maintains its "car-ness" autopoetically by imaging itself
06:22 < nsh> the separability of the imaged thing from the imagining mechanism is not as simple here
06:22 < nsh> *imaging
06:22 < nsh> *autopoietically
06:22 < nsh> as in dynamically self-creating
06:23 < nsh> self-recreating
06:23 < fltrz> I'm afraid I don't really understand the claim yet then
06:23 < nsh> that's fine :)
06:23 < fltrz> holographic principles are involved in say gene regulation?
06:23 < nsh> all of this took me several hundred discrete moments of insight-appreciation-realisation, none of which were especially trivial
06:23 < nsh> and i only half believe any of it due to lack of empirical conformation (or my not having found it yet)
06:24 < nsh> that's the thesis
06:24 < fltrz> I do appreciate an analogy between "neural networks" and "gene regulatory networks"
06:24 -!- Gurkenglas [~Gurkengla@unaffiliated/gurkenglas] has quit [Ping timeout: 255 seconds]
06:25 < nsh> okay, then try to imagine a kind of cascade process of information through a sequence of concentric spheres or deformed spheres, doing in both directions
06:25 < nsh> *going
06:25 < fltrz> and the binding affinities (DNA-RNA, DNA-protein, ...) are similar to weights in neural networks and give rise through an unknown fitness function to the fitness
06:26 < nsh> the information has a fixed minimal structure around which it rotates as it reaches the minimum size (which is at the DNA or RNA molecule)
06:26 < nsh> and as it expands outwards again it is modulated by what is going on in the cell as the wave passes through things
06:26 < nsh> at some point there is another reversal and the sphere of information begins contracting again
06:26 < nsh> (really this is happening continuously in both directions, like a spherical standing wave)
06:27 < nsh> ah anyway, i shouldn't ad lib this
06:27 < fltrz> :)
06:28 < nsh> Holographic Reduced Representations: Convolution Algebra for Compositional Distributed Representations - https://www.researchgate.net/profile/Tony_Plate/publication/2456333_Holographic_Reduced_Representations_Convolution_Algebra_for_Compositional_Distributed_Representations/links/0a85e53cad7b9d537f000000.pdf
06:28 < fltrz> don't worry about ad lib, as long as both parties understand its ad lib, it forces us to paraphrase, and between the 90% repetition and the 1% discovery is 9% paraphrasing
06:28  * nsh smiles
06:29 < nsh> but poorly phrased things can stick in the wrong places of the mind and obstruct better appreciation later. it's a double-edged sword
06:29 < nsh> as long as you understand that i am being terminologically imprecise and it should be taken as a kind of poetic telling
06:29 < fltrz> thats why it's important to understand when someone is going ad lib
06:29 < nsh> not a concrete and exact relating
06:29  * nsh nods
06:31 < nsh> Hyperdimensional Computing: An Introduction to Computing in Distributed Representation with High-Dimensional Random Vectors: http://www.rctn.org/vs265/kanerva09-hyperdimensional.pdf
06:33 < nsh> another way to think about it is this
06:33 < nsh> consider how you already create an (imperfect) correspondence between inside and outside when you just sit in a room as you are and model it 'inside' your neocortex
06:34 < nsh> how this is achieved is that the spatial location of abstracted features of the space you are sitting in is recreated in the cortex through the dynamics of cortical substructures which are assigned to features, subfeatures, etc.
06:34 < nsh> and which have their own transformational dynamic invariants fixed in by learning and the acquisition of object constancy as a cognitive faculty
06:35 < nsh> this allows the laptop and coffee cup inside my head to be continuously updates to be virtually coextensive with the coffee cup i can pick up and the keyboard on the laptop i have a typing feedback mechanism with
06:35 < nsh> so when i move my finger to the coffee cup and am about to feel its texture, i can anticipate this
06:36 < nsh> in fact i cannot not anticipate it because the neuronal assemblies are preemptively firing to facilitate the exact matching of comptuational simulated expectation of sensation with the percepts themselves arriving from my nerve endings
06:36 < nsh> these have ot match temporally for the cohesion of the external reality to be maintained
06:37 < nsh> this is a holographic klein inside-outside-correspondence phenomenon
06:37 < fltrz> well I understand that if we have a "redundant" basis, then an arbitrary random matrix to transform coordinates to a new random but still "redundant" basis incurs no loss of information
06:37 < nsh> we are only surprised when our internal simulation doesn't preemptively match the immanence of an external stimulation
06:37 < nsh> ie when we're tapped on the shoulder from behind
06:37 < nsh> exactly so!
06:38 < nsh> so there is a superabundance of redundancy of the bases as coordinate frame [bundles] and as the matrix and submatrix bases under tensor calculus that is being computer by the neural network
06:39 < nsh> from the physical side there is in fact no canonical non-local coordinate frame bundle, it's all spinorially transformable as long as twisting done along the nonlocal connection is undone again
06:39 < nsh> which is achieved by the bra-<kernel>-ket symmetry of spinor dynamics
06:39 < fltrz> there's been quite some progress in the last decades that showed that imaging through turbid media (say frosted glass) can be largely descattered (once the turbid medium has been characterized properly) to compute the original image
06:39 < nsh> but anyway, this means that over any distance there is an underdeterminism of coordinate frame orientation
06:40 < nsh> which is the same as redundancy in matrix basis
06:40 < nsh> right!
06:40 < nsh> in fact...
06:40 < nsh> this can result in 'super-resolution' phenomena
06:40 < nsh> you can see more detail by the convolution through a noisy medium if it can be computationally decoded
06:40 < nsh> this is now being harnessed in [soon to be] consumer optics and photography
06:41 < nsh> was recently use for some subwavelength stuff in astrophotography iirc
06:41 < nsh> roughly-stated again, you can beat the shot-limit and approach arbitrarily the heisenberg limit of metrological resolution
06:41 < nsh> *shotnoise limit
06:42 < nsh> .title https://www.youtube.com/watch?v=XV1Yu-X3Gzs
06:42 < yoleaux> Keith Motes - Linear Optical Quantum Metrology with Single Photons: Exploiting Spontaneously Genera - YouTube
06:42 < nsh> motes et al have some good papers and the references therefrom are worth pursuing
06:42 < nsh> this talk is a bit annoying because he had far too much fun with the mordor jokes
06:43 < nsh> well, for me anyway
06:43 < fltrz> does translation (rna to protein) occur in the cell nucleus too? or must RNA be exported for translation and then be imported again? if it can only happen outside the nucleus, then that's problematic for TAD domain / insulated neighborhood isolation, as we then really need either 1) domain walls 2) routing mechanism that returns the finished protein to the isolation domain where its RNA was transcribed
06:43 < nsh> .title https://arxiv.org/abs/1501.01067
06:43 < yoleaux> [1501.01067] Linear Optical Quantum Metrology with Single Photons: Exploiting Spontaneously Generated Entanglement to Beat the Shot-Noise Limit
06:44 < nsh> synthesis is usually in the cytoplasm
06:44 < nsh> but it's open
06:44 < nsh> .title https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1370363/
06:44 < yoleaux> Nuclear translation: What is the evidence?
06:44 < nsh> --
06:44 < nsh> Recently, several reports have been published in support of the idea that protein synthesis occurs in both the nucleus and the cytoplasm. This proposal has generated a great deal of excitement because, if true, it would mean that our thinking about the compartmentalization of cell functions would have to be re-evaluated. The significance and broad implications of this phenomenon require that the experimental evidence used to support it be carefully
06:44 < nsh> evaluated.
06:44 < nsh> --
06:45 < nsh> the assumptions of compartmentalisatoin are almost certainly reductive simplifications
06:45 < nsh> but they may hold more or less on average or when zoomed out enough
06:46 < nsh> .title http://jcb.rupress.org/content/197/1/7
06:46 < yoleaux> The enduring enigma of nuclear translation | JCB
06:46 < nsh> https://scholar.google.com/scholar?client=ubuntu&channel=fs&oe=utf-8&um=1&ie=UTF-8&lr&cites=11887090750567085449
06:47 < nsh> it's worth remembering that the eukaryotic cell evolved from the assimilation of the inside and something outside of the prokaryotic cell
06:47 < nsh> An inside-out origin for the eukaryotic cell - http://zero.sci-hub.tw/3638/0b9863650e0b35fd88526a93621571b8/baum2014.pdf#view=FitH
06:48 < fltrz> so even my "insulating neighborhood DNA loops as keychains for circularized mRNA" requires either a routing and addressing mechanism for proteins or else domain walls
06:48 < fltrz> unless there is nuclear translation going on
06:49 < fltrz> unless of course the genes regulation within insulating neighborhoods is not by DNA-binding proteins but by DNA-binding RNA
06:50 < nsh> well, what is a hologram when it's viewed? it's a collection of mutually interprenetrating dynamic 'domain walls' which is to say stationary =phase manifolds which then mutually interfere to create a coherent image of a static higher dimensional manifold
06:50 < nsh> i don't follow "unless there is nuclear translation going on", sorry
06:51 < nsh> i'm arguing that at least some informative factor of regulation is topological availability
06:51 < nsh> but this needn't be mediated by fixed hard [molecular] structures in every instance
06:51 < fltrz> I'm not seeing the hologram "force" that keeps transcription factors within their domain
06:52 < nsh> it would be necessary to show that the hologram steers molecules or increases/decreases the probability of interactions
06:52 < fltrz> nsh, we know that the insulating neighborhoods genes are *somehow* prevented from interacting with other insulating neighborhoods, but how can that happen without a membrane separating the neighborhoods?
06:52 < nsh> remember that protein binding sites require electronic interactions to be active
06:52 < nsh> and electronic interactions are mediated by spin waves
06:53 < nsh> if there is a convergence of phases that favour or rule out a certain kind of spin interaction that is necessary for a chemical interaction
06:53 < nsh> then that creates effective barriers without there being a stereochemical mediator
06:53 < nsh> and the fluid in which all of this is happening does not have on average random dynamics either
06:54 < nsh> well does have on average random dynamics. but they are not necessarily random at maximum detail
06:54 < nsh> they can be informed by radiative factors
06:54 < fltrz> then fenn proposed simply proximity to be what implements the isolation by slow diffusion, but if the RNA has to leave the nucleus to get translated into transcription factor, how does it nicely return to its proper insulating neighborhood?
06:55 < fltrz> so fenn's explanation could work, but only with immediate nuclear translation
06:55 < nsh> a thing to do maybe empirically is add light-emitting or light echoing tags bits of RNA and track them optically as them move about
06:55 < nsh> but i don't know if the tags would fuck up the dynamics
06:55 < nsh> at these scales you can't necessarily watch something without changing how it moves
06:55 < nsh> but might be able to do it 'weakly' by some cleverness
06:56 < fltrz> I guess if there are actual membranes, our best hope is the recent insta freezing into a glassy phase and then electron microscopy or smth
06:56 < nsh> ie you have a whole bunch of RNA molecules you can watch and you only look at any particular one a small fraction of the time then you try to computationally recombine these results to simulate having watch one continuously
06:56 < nsh> dunno
06:56 < nsh> i haven't kept up at all with what people can do
06:58 < fltrz> I remember reading about some kind of addressing/routing mechanism but I don't remember if it was for proteins or ... ?
07:00 < fltrz> if each insulating neighborhood had its own zip code, then perhaps the expressed protein can be properly returned...
07:01 < fltrz> I can't enlarge it because of dependency on javascript, but https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2829961/figure/A003889F4/ this reminds me of the keychain hypothesis
07:03 < fltrz> inset B that is\
07:06 < nsh> .head https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2829961/bin/cshperspect-NUC-003889_F4.jpg
07:06 < yoleaux> 200, image/jpeg, 126889 bytes
07:07 < nsh> .title https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5737799/
07:07 < yoleaux> Nuclear speckles: molecular organization, biological function and role in disease
07:07 < nsh> .title http://jcb.rupress.org/content/217/11/4025.short
07:07 < yoleaux> Mapping 3D genome organization relative to nuclear compartments using TSA-Seq as a cytological ruler | JCB
07:08 -!- CryptoDavid [uid14990@gateway/web/irccloud.com/x-rqyrgfzlrvnbaajt] has joined ##hplusroadmap
07:08 < nsh> .wik TSA-Seq
07:08 < yoleaux> "" — https://en.wikipedia.org/wiki/Secondary_Security_Screening_Selection
07:08 < nsh> nope
07:08 < nsh> .title https://www.sciencedaily.com/releases/2018/08/180828104040.htm
07:08 < yoleaux> Researchers develop 'cytological ruler' to build 3D map of human genome -- ScienceDaily
07:09 < nsh> "Yu Chen, Andrew Belmont, and colleagues from the University of Illinois at Urbana-Champaign have now developed a technique called tyramide signal amplification sequencing (TSA-Seq) that allows the distance of every gene from specific nuclear landmarks to be measured simultaneously."
07:09 < nsh> might be worth inquiring as to whether their datasets are available or can be made available
07:10 < nsh> "The TSA-Seq technique involves targeting an enzyme -- horseradish peroxidase -- to particular nuclear structures, such as the nuclear lamina that surrounds the nucleus or protein-containing granules called nuclear speckles that tend to be found in the center of the nucleus. The horseradish peroxidase then generates a highly reactive molecule called tyramide that can be used to label any DNA in the enzyme's vicinity. The closer a gene is to the enzyme,
07:10 < nsh> the more it will be labeled. So, when researchers subsequently sequence the cells' DNA, they can calculate how close each gene was to the nuclear structure tagged with horseradish peroxidase."
07:14 < fltrz> ok, I don't know what those speckles are, so I'll have to read up on that first
07:15 -!- Gurkenglas [~Gurkengla@unaffiliated/gurkenglas] has joined ##hplusroadmap
07:16 < fltrz> "Although the interior of the nucleus does not contain any membrane-bound subcompartments, its contents are not uniform, and a number of nuclear bodies exist, made up of unique proteins, RNA molecules, and particular parts of the chromosomes."
07:17 < fltrz> so there are *NO* subcompartments!
07:18 < fltrz> its strange that conventionally there is no nuclear translation, but the ribosomes are generated in the nucleoli within the nucleus, and then exported
07:20 < fltrz> so it seems like in the cytoplasm translation of mRNA is independant of mRNA origin neighborhood, but within the nuclear bulk there is no translation allowed, to me this suggests that IF there is translation in the nucleus, it is restricted to happen WITHIN the nucleoli
07:21 < fltrz> so if each insulating neighborhood had its own one or more nucleolus/nucleoli then translation might result in the resultant protein to be released within the original insulating neighborhood
07:22 < fltrz> i.e. RNA of neighborhood x goes to a nucleolus belonging to neighborhood x, and once translated is released back into neighborhood x, and the same for neighborhoods y and z, such that there is no or little crosstalk
07:24 < fltrz> this might explain why translation is hard to detect in the nucleus, if ribosomes are only allowed to pass the bulk (from a nucleolus to be exported to cytoplasm) in a form or state where it can not translate mRNA for that would allow neighborhood crosstalk to occur
07:27 < fltrz> this also suggests that those "global" proteins that need to be able to cross insulating domains are those for which the nuclear membrane are transparent
07:33 < fltrz> I'm still dissatisfied with a purely diffusional isolation mechanism, unless there is a high breakdown rate for protein, even if 99% of expressed proteins end up in their correct insulating domain, those 1% will build up over time and cause a uniform distribution
07:46 -!- Gurkenglas [~Gurkengla@unaffiliated/gurkenglas] has quit [Remote host closed the connection]
07:46 -!- Gurkenglas [~Gurkengla@unaffiliated/gurkenglas] has joined ##hplusroadmap
07:47 < fltrz> oh there is only one nucleolus? then we need routing again.. sigh
07:51 < fltrz> "Although usually only one or two nucleoli can be seen, a diploid human cell has ten nucleolus organizer regions (NORs) and could have more nucleoli. Most often multiple NORs participate in each nucleolus."
07:55 < fltrz> so instead of each insulating domain their own nucleolus, each insulating domain their own NOR?
07:57 -!- yashgaroth [~yashgarot@2606:6000:c308:f700:8ca2:15a0:7664:ffb2] has joined ##hplusroadmap
08:05 < nsh> you seem super interested in one-to-one correspondences. i expect this tendency will result in disappointments :)
08:15 < fltrz> it needs a strong correspondence to explain the insulating domain, we observe the domain boundary markers and observe the isolation effect, but we lack the mechanism that constrains the expressed transcription factor to within the domain
08:16 < fltrz> so yes I am looking for the isolation mechanism which effects the 1 on 1 source neighborhood to target neighborhood mapping
08:16 < fltrz> perhaps routing, perhaps local translation, perhaps...
08:17 -!- spaceangel [~spaceange@ip-86-49-16-65.net.upcbroadband.cz] has joined ##hplusroadmap
08:19 < fltrz> imagine reading about an observation that every time you eat a cookie you get mail,... then surely, even though the knowledge that it is so is satisfying, you also want to know *how or why* every time you eat a cookie you get mail
08:21 < fltrz> remove the insulating domain boundary marker and you get a defect, the genes interact predominantly within insulating domains, but a gene can not directly influence another gene, it needs to be "expressed" = transcription + translation; but at the same time translation supposedly can not happen locally within the domain, so when the protein is finally formed it needs to end up in its original domain again!
08:30 < fltrz> so not only are their supposedly no subcompartment membranes, there is supposedly no nuclear translation either, ... that necessitates a routing mechanism for proteins and an addressing scheme for insulating domains, but even if there is such routing and addressing, how does removing the boundary marker affect the routing / addressing?
08:31 < nsh> but a 1-to-1 correspondence followed by a later 1-to-1 correspondence does not need a continuously maintained correspondence in between
08:31 < fltrz> so it still seems like there is some kind of protein-DNA adhesion force (regardless of sequence affinity) that constrains protein movement to the DNA sequence, in such a way that the protein while walking along the DNA can not cross the boundary markers
08:31 < nsh> think about the path integral formulation of a photon emission and absorption event...
08:32 < nsh> it can employ a very large intermediary space where it is nonlocal and then still be atomic at 'the end'
08:34 < fltrz> the difference is that for a lense we *know* the effect of lense refractive index and thickness such that we understand that light emission in an object plane will focus to a point in the image plane
08:34 < fltrz> we understand the "routing and addressing" of photons by lenses
08:35 < fltrz> but we don't understand the mechanism for cells.
08:36 < fltrz> I'm not claiming the ultimate 1-on-1 correspondence is impossible for cells, I'm saying we (or at least I) don't understand the underlying mechanism for isolated gene expression
08:37 < fltrz> to make your analogy more complete would be to take the example of a lens *but without any understanding of optical path length, refractive index, interference and linear superposition*
08:38 < fltrz> and in that case it is mandatory to explain how a lens succeeds in mapping each point on the object plane to a single point in the image plane
08:41 < fltrz> I probably just have to read this: https://en.wikipedia.org/wiki/Protein_targeting
08:45 < fltrz> but even targeting is insufficient to explain the boundary effect of the boundary markers (which create locked loops of DNA with CTCF and cohesin)
08:46 < fltrz> so on top of targeting we need a general (nonsequence specific) attraction to DNA of the protein, and some inability to cross CTCF / cohesin which are form the endpoints of the DNA loop
08:49 < fltrz> unless one of the assumptions is wrong (i.e. there are membranes either separating the insulating neighborhoods, or forming "soap bubble blower layer" on the loop, OR there must be nuclear translation, and this nuclear translation is happening privately within each insulating domain, but even then the effect of the boundary markers is unexplained)
08:50 < fltrz> OR the insulating neighborhoods are a myth?
08:52 < fltrz> why give each insulating neighborhood its own zip code, if you can simply use longer DNA-binding site sequences to prevent crosstalk?
08:53 < fltrz> what is going on?!
08:54 < fltrz> perhaps there is a thermodynamic limit on DNA-binding site sequence length which forces to use 2 sequences: 1 for rough targeting to the insulated neighborhood and 1 for the precise operator?
08:56 < nsh> fltrz, right
08:56 < nsh> (general agreement to the above)
09:28 < fltrz> could the insulating neighborhood "boundary markers" or "anchors" be some kind of locking to keep chromatin either tightly locked or forcibly unlocked at a longer radius than the histones?
09:29 < fltrz> does euchromatin even have histones or is it only for heterochromatin?
09:30 < fltrz> i'm also still confused what happens with the complementary DNA strand regarding these anchored loops
09:34 -!- kuldeep [~kuldeep@unaffiliated/kuldeepdhaka] has quit [Excess Flood]
09:35 -!- kuldeep [~kuldeep@unaffiliated/kuldeepdhaka] has joined ##hplusroadmap
09:53 -!- Urchin [~urchin@unaffiliated/urchin] has joined ##hplusroadmap
10:02 < fltrz> NLS and importins https://en.wikipedia.org/wiki/Nuclear_localization_signal
10:04 < fltrz> for the bipartite sequences the short "relevant sequences" are separated by a 10 amino acid "spacer", perhaps the short sequences are the localization sequence for the *nucleus* and the spacers are in fact the more precise insulating neighborhood zip codes?
10:16 -!- l_wl [~l_wl@pool-108-45-141-95.washdc.fios.verizon.net] has quit [Remote host closed the connection]
10:17 -!- l_wl [~l_wl@pool-108-45-141-95.washdc.fios.verizon.net] has joined ##hplusroadmap
10:19 -!- yashgaroth_ [~yashgarot@2606:6000:c308:f700:8ca2:15a0:7664:ffb2] has joined ##hplusroadmap
10:21 < fltrz> what if in fact there is no isolation at all caused by the insulating neighborhood anchors, and that the correct positioning and binding of the anchors is merely required to correctly place the enhancer in the vicinity of its target? that would explain the defects of anchor absence/mutation without any gene regulatory isolation at all!
10:22 -!- yashgaroth [~yashgarot@2606:6000:c308:f700:8ca2:15a0:7664:ffb2] has quit [Ping timeout: 264 seconds]
10:23 < fltrz> then the concentrations are approximately uniform within the cell nucleus and normal gillespie algorithm would apply
10:24 < fltrz> no need for diffusion effect, nor for membranes etc
10:38 -!- Urchin [~urchin@unaffiliated/urchin] has quit [Ping timeout: 244 seconds]
10:55 -!- yashgaroth_ [~yashgarot@2606:6000:c308:f700:8ca2:15a0:7664:ffb2] has quit [Quit: Leaving]
10:56 -!- yashgaroth [~yashgarot@2606:6000:c308:f700:8ca2:15a0:7664:ffb2] has joined ##hplusroadmap
11:32 -!- Urchin [~urchin@unaffiliated/urchin] has joined ##hplusroadmap
11:58 -!- Gurkenglas [~Gurkengla@unaffiliated/gurkenglas] has quit [Ping timeout: 255 seconds]
13:18 -!- FourFire [~FourFire@234.103.189.109.customer.cdi.no] has joined ##hplusroadmap
14:33 -!- justanotheruser [~justanoth@unaffiliated/justanotheruser] has quit [Read error: Connection reset by peer]
14:47 -!- justanotheruser [~justanoth@unaffiliated/justanotheruser] has joined ##hplusroadmap
15:22 -!- spaceangel [~spaceange@ip-86-49-16-65.net.upcbroadband.cz] has quit [Ping timeout: 240 seconds]
17:13 -!- ebowden [~ebowden@unaffiliated/ebowden] has quit [Remote host closed the connection]
17:15 -!- ebowden [~ebowden@unaffiliated/ebowden] has joined ##hplusroadmap
18:07 -!- helleshin [~talinck@cpe-174-97-113-184.cinci.res.rr.com] has joined ##hplusroadmap
18:15 -!- helleshin [~talinck@cpe-174-97-113-184.cinci.res.rr.com] has quit [Read error: Connection reset by peer]
18:19 -!- helleshin [~talinck@cpe-174-97-113-184.cinci.res.rr.com] has joined ##hplusroadmap
19:13 -!- ebowden [~ebowden@unaffiliated/ebowden] has quit [Ping timeout: 246 seconds]
19:23 -!- ebowden [~ebowden@unaffiliated/ebowden] has joined ##hplusroadmap
19:27 -!- CryptoDavid [uid14990@gateway/web/irccloud.com/x-rqyrgfzlrvnbaajt] has quit [Quit: Connection closed for inactivity]
19:59 < kanzure> hmph
20:28 -!- strages [uid11297@gateway/web/irccloud.com/x-zyafuzhkxzfpvsiv] has joined ##hplusroadmap
21:56 -!- yashgaroth [~yashgarot@2606:6000:c308:f700:8ca2:15a0:7664:ffb2] has quit [Quit: Leaving]
22:38 -!- Urchin [~urchin@unaffiliated/urchin] has quit [Remote host closed the connection]
22:44 -!- Gurkenglas [~Gurkengla@unaffiliated/gurkenglas] has joined ##hplusroadmap
22:53 -!- Gurkenglas [~Gurkengla@unaffiliated/gurkenglas] has quit [Ping timeout: 255 seconds]
23:49 -!- strages [uid11297@gateway/web/irccloud.com/x-zyafuzhkxzfpvsiv] has quit [Quit: Connection closed for inactivity]
--- Log closed Sun Mar 03 00:00:05 2019