Copy Signed Initials with airSlate SignNow

Copy signed initials

right in today's video we're going to go over how to do the initial CTF estimation which you would want to do even prior to doing template matching although template matching doesn't require CTF corrected input since we nearly always filter by low-pass to a resolution lower than the first CTF zero this initial estimate gives us a good first objective look at the quality of the data and allows us to eliminate tilt series that may otherwise have wasted our time later on so we are currently in the tutorial data set that we started before in the fixed Stax directory and there's one file that I forgot to mention as required when I did the initial setup and course alignment with iMod and it was this one called our pre fixed tilt one order and all this is is a list of angles that represent the port of the data were collected in and this could be anything like a bidirectional from zero by direction from 20 you go from negative sixty to sixty or do something like the dissymmetric tilt scheme all that matters is that you list the tilt orders in the order they were collected so I'm going to show you one way to do it there's sure many better ways and if any of you have good suggestions I would love for you to post them so we can include that for a by Dec bi-directional tilt scheme like we have here one simple way to do it is to open your tilt angle file until one dot till it's going to text editor and this was collected from positive twenty to negative sixty so I'm going to take the other angles and delete them or cut them ctrl X and make sure there's no spaces we want to save this as tilt one order make sure you don't accidentally overwrite your tilt file that has your tilt angles in it now we're going to save make a new file in this case I'm going to add ctrl n and paste the values we cut out and we can't use that directly but we can do we're just going to save it as temp text and we're good to go so I'm close both days now if we print out temp text we could just append this to our other file but what we want to do is invert the order so when you can do as with the shell command sort if we go sort temp text we see they come out inverted order now if you have negative numbers you may need to include the G flag which just tells it that it's numerical input and treated as such and you can reverse the order that you sort things to by including an R flag which we don't need to use in this case but just for complete description here so what we're going to do is we have our tool on that order that lists our angles from our starting increment 20 which in this case it's 20 once it's the refined tilt angles from the I'm on alignment all the way up to negative 60 what we'll do is we're going to sort our temp text we're going to do that if we use two little caret that means a pen - if you do one care it will overwrite the file all right so one ducktail it's all that does is it prints out and yeah prints out what we had in temp dot txt in reverse order and it then puts it in there so if we print out to one that tilt now we see that we've attacked on these values so we go from neg positive 20 to negative 60 we jump pulley back to positive 24 and work our way up to positive 60 so we now have our tempt X or tilt one order and to remove temp tags it's probably good just to check to make sure giving the sort of things I like to that we didn't screw up the angles in tempted oh no Drita all right so now we have the order of data collection was important for using exposure filter and we're going to go back and from our main project directory when you're on a.m. clarity CTF estimate and we want to just give it the prefix for the one we're concerning which is tilt one and you'll eventually then do tilt two and then the last parameter is just to give it a specific GPU most of the programs in iam clarity divide the jobs up over a certain range or certain set of GPS that you give it to work with and it does it automatically but since we aren't fully into the iam clarity workflow here you can specify exactly which GP you want it to use which MATLAB indexes from one whereas nvidia indexes from zero so i'm going to send that off and we'll wait a minute and get the return and we get back an error because i forgot to include the parameter file which is good because in this tree recording i forgot to even mention it so what i've done there is a parameter file that specifies the number of things that we have to define for each project and we get a copy of that whenever we download iam clarity so what you want to do is copy wherever you have the enquiry installed in the docs folder there's something called master Quran you just want to copy that to your current working directory and rename it so on the copy here is tutorial m and then we want to change that to suit argument project so we'll go over a little bit what's in the tutorial for in the parameter file so this consume daunting at first but really there's a lot in here that you don't change too frequently so the first thing is selecting a project name which I'm just calling relying tutorial here can be anything you want there can't be spaces and ask to start with a letter not a number and that's good I think you can only use underscores in the names that has to do with our MATLAB treats structures we've also defined the number of GPS available to us on my test rig I only have one GPU if you have more than one GPU you can specify it as say four and that will select up to four GPUs or we can provide a list like GPUs one and three the average memory parameter you're just going to leave as one when you get to full sampling lay in the process sometimes you can change to two to conserve some memory but in most cases you're going to leave that some one the cetf parameters that we're going to change now at the very bottom of the parameter file or just about the very bottom and the 3d CTF is something that we're currently implementing and it's working very nicely but it's not something you would do till later in the refinement stage for now we're going to work just with the two-dimensional CTS so the microscope parameters first to define our pixel size and everything in the section is in meters and volts so some people use nanometers some programs use angstroms we just like to keep it nice and consistent so use SI units so it's two point one seven angstroms is our pixel size of the data set it's not a super resolution set which we'll talk about in a different tutorial the spherical aberration is two point seven militants use does three hundred kilovolts and amplitude contrast is anywhere between point O seven and 0.14 typically in a lot of papers that i've seen from the Biggs lab he likes to use point one it's kind of a hit and miss thing I just settle for something sort of in the middle the search range that you use for the defocus is pretty robust errors at this point other programs are pretty sensitive to using too wider range but we're just going to specify four as the middle and go plus and minus two and a half microns with a final cutoff of nine angstroms it automatically determines where the peak after the first zero is and then it scales back to the first inflection point from there it's good to choose a to focus cutoff that includes as much useful information as possible but not a ton of noise and you're not really going to know how that is for your data set until you run it so around eight or nine is a good starting point you may need to bump that up or down a couple angstroms the cumin of electron dose is eighty electrons brings from squared for this data set in our lab typically we've been using closer to 100 or 120 this allows you to maximize the low frequency information that you can use to improve the alignment of both the tilt series and when you want to refine the actual particle positions of the projections while at the same time preserving high resolution information and we use the optimal exposure filter that to granny co-published so the bead diameter refers to the diameter of the fiducial markers which in our case is 10 nanometers and then there's some astigmatism parameters which we'll we'll call those advanced parameters from now this is a wide enough range that you probably shouldn't have to change it if for some reason you want to turn it off you can change that boolean to a zero otherwise this is pretty robust as well and for this particular data set I can't remember the number off my top of my head but it's pretty small about 100 nanometers to focus or a hundred angstroms sorry the number of workers only specifies in this case when you're interpolating so we use a spline based interpolation kernel which if you're using the super resolution data set you're probably going to want to limit it because we can get pretty huge pretty fast but it's not particularly slow even with a few workers so keeping them below isn't going to hurt anything the Delta Z tolerance is in in our case 25 to 50 nanometers typically that's the change in anticipated defocus that you allow into the averaging and this usually gives very good results if you have really low sevilla noise data like you're working with six cells you may want to increase this which will give you a lower resolution estimate but it will give you more power in the signal which is what you would want for that situation anyhow you don't expect to me to get past 15 angstroms if you've got a 300 nanometer thick cell which would obviously have to be live stir milled to be that thing these parameters we're not going to worry about this tutorial if it discussed in the documentation okay so that's a brief overview of the tutorial parameters we're going to want to use and now we'll run that again except we'll go Ian clarity CTF estimate will include the name of the parameter file tutorial M the prefix or the tilt series and the GPU ID and we'll check back in a few minutes alright now this will print out a bunch of stuff to your terminal if you're running interactively on your own desktop and we'll end with the astigmatism search so the log files that iam clarity saves are hidden to create it or prevent clutter which you can show them if you do LS with the a tag they'll show hidden files would start with a dot so this is the log file it contains everything that just print it out on your screen in most cases you aren't going to probably need to look at that but that will help troubleshooting if you run into bugs down the road so first we're going to look at what to stab in here so under fix tax we run CTF estimate if it doesn't already exist a new directory is created called CTF which we're going to move into and then a handful of files which are prefixed by our tilt series prefix are put out so we'll start with the astigmatism dot text just because we don't need to talk too much about it this just contains some files anyone check and make sure it's reasonable so this was an astigmatism of 37 nanometers which is pretty small it's not tiny in the single particle world it's actually pretty big but for tomography it's not too bad it isn't really until you start getting up into the hundreds of nanometers that it really becomes essential to correct and even then it's only going to make a big difference when you're around a hit angstrom resolution so the other files we've put out that are of interest instead of describing all these which I'll do in the docs I'm just going to show you the most relevant for now is this PS radial dot PDF and what that gives you is a fit of the actual theoretical curve which is here in green which you can see has an envelope applied to it and the power spectrum which is here in blue now we can see that we have very strong signal the whole way out to name nine angstroms which is what I fit so it probably means that we actually wanted to include more data one of the reasons we have such strong signals we've got a ton of ribosomes which are obviously very strong electron scatters and the ice is reasonably thin for the specimen to the sake of the tutorial I'm not going to redo this but I would recommend that if you were running this on your own project and we saw this high power all the way out to your limit I would stretch it out now remember this doesn't not linear so from nine I would probably go to seven and see what you get from that for now we're going to stick with the value that is negative three point six six mic on as the defocus and that also created this file that ends with tilt just like our righto file but instead of just containing information about the tilt angles there's a bunch more but you don't really need to worry about what's in there it looks like a mess this is a precursor to the geometry that's stored in the binary files I am clarity uses so to make sure not to edit this but I do want to point out this Ally one so that is going to be prefixed in a can change of the experiment based on whether or not you've done any tomogram particle polishing and for every time you do to him again particle polishing that number will increment by one we'll talk about that in its own video basically these files are here you can explore these other ones which are all centered around troubleshooting and assessing the quality of the fit but we can see already just from the radial power spectra that things look good so from there we've completed the CTF estimation one other things happened while we were running is this new directory project called alive stacks so what that is is whenever you do the CTF estimation it is one of the ways I am clarity will use to apply the transformations we found an eye mod or layer that we were refining the in clarity to create an align stack so I'm going to open that at a binning of two once before just to make it quick and all this is is instead of now being tilt one not fixed it's called tilt one underscore allied out fixed it's the same tilt series we had but now it's with the alignment that we had an eye mod which shouldn't look very different because this was already a pretty long stack so now that we've got this file and you should do this also for tilt number two we'll be able to move on to actually setting up areas that we want to reconstruct from there moving on to template matching all right thanks