|
Superhydrophobicbosssuggestsbutitsatrocious?
|
# ? Oct 16, 2019 07:16 |
|
|
# ? Jun 10, 2024 13:10 |
|
ascii genitals posted:I don't do a lot of lcms but ab sciex ms with an Agilent lc seems like what you are looking for. Just spent the afternoon with a LC MS operator on a Sciex MS and the interface was very much clicking with the mouse, which gets old fast with 1000+ compounds. Anyone here got experience with setting up automated analysis on a Sciex that can point me to some good starting point for that?
|
# ? Oct 16, 2019 17:19 |
|
Sundae posted:Superhydrophobicbosssuggestsbutitsatrocious?
|
# ? Oct 16, 2019 22:36 |
|
Aaaaaand we just had a sterility failure. Time for a three month investigation, at a minimum....
|
# ? Oct 18, 2019 17:17 |
|
My last sterility failure took nine months to resolve and I'm going to be dealing with the fallout for the rest of my life.
|
# ? Oct 18, 2019 21:18 |
|
Cardiac posted:Just spent the afternoon with a LC MS operator on a Sciex MS and the interface was very much clicking with the mouse, which gets old fast with 1000+ compounds. What part are you trying to automate? Same analytical method with different compounds, trying to automate the identification? Same compounds different concentrations?
|
# ? Oct 23, 2019 15:49 |
|
ascii genitals posted:What part are you trying to automate? Same analytical method with different compounds, trying to automate the identification? Same compounds different concentrations? I basically need a way to handle 1000-10000 different known compounds that come in mixes of 25-50 without having to do manual copy pastes or analysis. Chromatography or MS is nothing advanced, it is just that analysis software is not designed for this. Ideally I would want to design batch automation and MS methods in text files using Python/whatever and then having some form of automatic analysis for the chromatograms, since it in 90-95% of the times is a single peak. The only thing I actually care about is the retention time and some information on peak shape.
|
# ? Oct 23, 2019 17:41 |
|
Generally you can set a data system to spit out integration RT, abundance, and some other information as excel (csv or better yet tsv.) I don't know Analyst software so I don't know how to recommend the exact steps. Is this lcms single quad, tandem quad, or some accurate mass stuff? Keep in mind lcms sq libraries of spectra dont really exist like it would for gcms (because the spectra is dependant on ion optics that will differ instrument to instrument.) If it is ms/ms or accurate mass you can identify based on mrm or product ion scan, accurate mass can go by fragment masses. You'd need to find either the MRM / MS/MS information, or the exact mass info, for all your compounds or acquire it yourself. Once you run it on your lc method youll also have the RT. For automation you really need the precise RT as well as the MS specific info. Once you figure this stuff out you may find you want to automate creation of the sequence to acquire the data. This can usually be done by exporting a csv or some other excel based method of generating a sample list.
|
# ? Oct 23, 2019 18:24 |
|
ascii genitals posted:Generally you can set a data system to spit out integration RT, abundance, and some other information as excel (csv or better yet tsv.) I don't know Analyst software so I don't know how to recommend the exact steps. We are running on both single and triple quads. Our own library is tuned by MRM, while client libraries are not. We know the monoisotopic masses of what we inject and basically want to detect the RT for them. The RT will vary depending on column and the difference between different columns is the interesting value. I guess I will have to go digging into how the text files are formatted to figure out a way.
|
# ? Oct 23, 2019 18:33 |
|
Cardiac posted:We are running on both single and triple quads. Our own library is tuned by MRM, while client libraries are not. Is the challenge you want to avoid manually defining RTs? Could you just make a method with RT windows spanning the whole run, export, and automate the rest from there? I only know chromatography software, so not sure if that makes sense with MS
|
# ? Oct 23, 2019 20:57 |
|
Epitope posted:Is the challenge you want to avoid manually defining RTs? Could you just make a method with RT windows spanning the whole run, export, and automate the rest from there? I only know chromatography software, so not sure if that makes sense with MS Pretty much. Chromatography and MS is pretty insistent on having the correct mass at the correct RT, which does not apply in this application. Your idea is something I have already been thinking about so that makes sense. Now to figure out how that works on our machine. Alternatively I was thinking about converting the MS data to an open format and try a OpenMS/KNIME workflow. Edit: now to find out how to go from text file to acquisition method in analyst, cause having to manually create 100 of them is going to be tedious. Creating a batch acquisition is easy. Cardiac fucked around with this message at 17:27 on Oct 24, 2019 |
# ? Oct 24, 2019 04:34 |
|
If you define the RT of a compound as the entire run time, meaning you set the MS to look for ions of a mass characteristic to your compound (MRMs, SIM, product ion scan whatever), you will limit the number of total compounds you can analyze in a single run. Unlike GC or LC there is an opportunity cost (time) associated with detection. This is why triggered MRM exists, you look for one high abundance MRM for a compound and if you get a hit, start looking for additional confirmatory MRMs for a few seconds. Unfortunately for gcms the peak shape is sharper so there is less time to detect the primary mrm and trigger the secondary MRMs while still getting sufficient sampling rate... I have no actual experience using tMRM but I know it exists to try and optimize the number of MRMs that can be run at one time. There's no free lunch though: tMRM with have some consequences on performance for lcms for the reason mentioned above. Screening for hundreds of compounds can be done, but you need to have the RTs in order to narrow the time range where you are attempting to detect a compound to something like RT +/- 0.2 minutes. In this case it would be realistic to screen for 200-250 compounds in a 20 minute chromatographic run. I have seen that 20 min run get optimized down to 5 or 10 min, but you wouldn't just start with that. If you can get it down to a 5 minute run you could screen for 1000 compounds by rerunning the same sample multiple times with different methods, but this would still just be screening no quantitation. ascii genitals fucked around with this message at 20:22 on Oct 24, 2019 |
# ? Oct 24, 2019 20:16 |
|
So after months of haggling, our lab is getting a gram staining machine so we don't have to do it manually anymore. But I'm curious if it actually makes a big difference in terms of time, staining quality etc. Especially because I'm pretty sure you still have to manually pipette your samples onto slides and wait for them to dry, which IMO always takes longer than the bit with all the chemicals anyway. (I admit it'll be nice to not worry about accidentally loving up the decolorizer anymore, lol.) Has anyone in here used one of those bad boys?
|
# ? Oct 25, 2019 01:40 |
|
Question for instrument service goons, how did you guys get your foot in the door? After
|
# ? Oct 25, 2019 13:35 |
|
Shrieking Muppet posted:Question for instrument service goons, how did you guys get your foot in the door? After I'd say to target the manufacturers of instruments that you've worked with (and fixed?) to start out with. Reach out to the FSEs that you've worked with in the past and ask them about their jobs. If you're generally helpful and saved them trips or do a lot of the troubleshooting they'll remember you and would probably refer you if their team is hiring. Otherwise just apply to the roles and be ready to talk about how you've troubleshot and fixed anything. Cars, lab equipment, bikes, Arduino stuff you've made, it's all fair game here and they'll be looking to see something that demonstrates you can turn a screwdriver or use an ammeter.
|
# ? Oct 25, 2019 14:25 |
|
^^ that's probably better advice Shrieking Muppet posted:Question for instrument service goons, how did you guys get your foot in the door? After Found their booth at a conference and chatted for a while, they actually remembered me and reached out later that year when a FSE position opened up. I'm still a bit shocked it worked out. It was a good gig but 5 years of traveling was way enough for me.
|
# ? Oct 25, 2019 19:03 |
|
Shrieking Muppet posted:Question for instrument service goons, how did you guys get your foot in the door? After
|
# ? Oct 26, 2019 02:31 |
|
Dik Hz posted:Stuff like that. The churn rate on FSEs is so high that jobs come open frequently. Must be traveling that kills it. Got asked about my interest in a sales position for a instrument manufacturer in my field. Cold calling was one thing, but 50-75% travel time killed any interest. As for my chromatography question, I think I have the solutions. Batch list is easy, since it is a text file. New acquisition method for each mix means learning Visual Basic apparently. We are running SIM so the masses are known and the wiff files can be converted using ProteoWizard and read into KNIME/OpenMS for processing.
|
# ? Oct 26, 2019 05:10 |
|
Yep it's the travel. Nearly everyone eventually moves to another role or changes careers. I had a couple coworkers leave and go teach or write a book---burned out entirely. It is a great job though. You make tons of connections, hone your people skills, etc etc.
|
# ? Oct 26, 2019 17:45 |
|
More acquisition softwares need a well documented api/sdk. Automating acq method a and sequence file creation is a hot topic. I recently worked on a MRM optimizer that automatically creates product ion scan and mrm methods, but it's for method development not actually collecting new data. ascii genitals fucked around with this message at 17:49 on Oct 26, 2019 |
# ? Oct 26, 2019 17:47 |
|
My purified water system has no water and industrial water is pouring out of my HEPA filters. Happy loving friday, lab rats.
|
# ? Nov 8, 2019 19:53 |
|
Sundae posted:My purified water system has no water and industrial water is pouring out of my HEPA filters. Happy loving friday, lab rats. Wouldn’t water ran through a HEPA filter be considered purified? Just not in the way you want?
|
# ? Nov 10, 2019 22:27 |
|
Shrieking Muppet posted:Wouldn’t water ran through a HEPA filter be considered purified? Just not in the way you want? How many validation studies have we done on this new water system? I'm expecting at least an oq / pq out of this if we're gonna release the system to production.
|
# ? Nov 10, 2019 22:55 |
|
My analytical lab manager gave his two weeks' notice today. Anyone want to run an analytical lab focused on polymer characterization for me? I guarantee the lab has more bench space than any lab you've ever worked in before.
|
# ? Nov 15, 2019 03:28 |
|
Dik Hz posted:My analytical lab manager gave his two weeks' notice today. Anyone want to run an analytical lab focused on polymer characterization for me? I guarantee the lab has more bench space than any lab you've ever worked in before. Where about?
|
# ? Nov 15, 2019 13:58 |
|
Shrieking Muppet posted:Where about?
|
# ? Nov 17, 2019 02:13 |
|
Now I'm really curious, I've worked for a ton of companies in that area. Hmmm so much bench space... Another clue?? Anyone who is interested: it is a very nice area and he is right, cost of living is amazing compared to what you may be used to.
|
# ? Nov 17, 2019 03:02 |
|
$80k for a lab manager position seems...
|
# ? Nov 17, 2019 04:43 |
|
ascii genitals posted:Now I'm really curious, I've worked for a ton of companies in that area. Hmmm so much bench space... Another clue?? Lyon posted:$80k for a lab manager position seems... Dik Hz fucked around with this message at 19:08 on Nov 17, 2019 |
# ? Nov 17, 2019 19:04 |
Lyon posted:$80k for a lab manager position seems... ...good for the research triangle.
|
|
# ? Nov 18, 2019 14:28 |
|
That Works posted:...good for the research triangle.
|
# ? Nov 19, 2019 00:48 |
|
Well, since our building is getting demolished, i will not be doing experiments in the lab again for several months.
|
# ? Nov 19, 2019 00:56 |
Dik Hz posted:It's Piedmont Triad, not Research Triangle, so the money goes a little further. Oh duh, my bad
|
|
# ? Nov 19, 2019 02:37 |
|
Dik Hz posted:My analytical lab manager gave his two weeks' notice today. Anyone want to run an analytical lab focused on polymer characterization for me? I guarantee the lab has more bench space than any lab you've ever worked in before. Free bench space? More like free to quality more instruments.
|
# ? Nov 19, 2019 03:06 |
|
Mustached Demon posted:Free bench space? More like free to quality more instruments. Dik Hz fucked around with this message at 05:29 on Nov 19, 2019 |
# ? Nov 19, 2019 05:24 |
|
Dik Hz posted:I put in over 1200 square feet of bench space. I just went from six to midnight, drat!
|
# ? Nov 19, 2019 12:21 |
|
Anyone here know whether Agilent provides an interface that allows automatising batch runs and acquisition methods for MS? I know Sciex provides a VB.net that should allow for this. In contrast, Thermos software does not offer this alternative. Googled a bit on Agilent, but no luck. Analysis is less of an issue, as Proteowizard can convert from vendor formats to mzML and the rest can be solved by programming.
|
# ? Nov 20, 2019 07:51 |
|
A member of our data review group is trying to tell me my solution preparation is wrong because a volumetric flask is less accurate than a graduated cylinder.
|
# ? Nov 22, 2019 19:25 |
Shrieking Muppet posted:A member of our data review group is trying to tell me my solution preparation is wrong because a volumetric flask is less accurate than a graduated cylinder.
|
|
# ? Nov 22, 2019 19:32 |
|
|
# ? Jun 10, 2024 13:10 |
|
Hell yeah
|
# ? Nov 22, 2019 19:52 |