Using LabArchives Electronic Notebooks to Support Research Data Management Plans (RDMPs)
Transcript
All right, so let's talk about good research data management and sharing plans. First up, if you are here today and don't know what a data management and sharing plan is, these are basically formal documents that describe what you do, how you do, and how you document your work. If you don't have a data management and sharing plan, the first thing I would recommend is check with your funder.
So as an example, NIH publishes excellent templates that are available on the various NIH websites for specific information, and these actually need to be filed with NIH. But I would also volunteer your excellent libraries on campus and your research data management folks who can give you some guidance on good data management practices, good practices in your particular area of research, not just, you know, broad, but more kind of your discipline, but also how the data management and sharing plan affects your whole life cycle of your data. So from coming up with an idea to documenting, collecting your data, to analyzing it, to preparing figures and preparing your publication, all the way until looking for repositories, publishing of data, public access of your work.
So kind of that full research life cycle. Now you'll often hear in the industry an acronym of FAIR, findable, accessible, interoperable, and reusable. So what supports findable data? Search is the simple answer, but in LabArchives we have great search.
You can type in a term, you can click the drop down and see our advanced search tools. So you can search by notebook, by user, by guest, by entry type, by page title, and look for the data that you need. Now the problem is your search is only as good as your data.
So if your team is just uploading files that are called untitled one, or using really inconsistent page names, really inconsistent folder names, or you have no consistency at all, right? Day-to-day your documentation is going to be different. It's harder to search for your work. Now there are plenty of ways to help improve search and we're going to talk about it.
Accessible is a pretty involved process talking about both access, being able to get to your notebooks, even if you're working remotely, but also permissioning. We'll talk about that as well. Interoperable is all about being able to use your data in the long term, right? Sure, you might develop this code, this script, this work that works great on a specific version of Java, a specific version of PHP, a specific version of Chrome on a Windows computer, but you should also make sure it works on other operating systems, it works in other systems, and will be usable for the long term.
Again, this is one of those areas where I would plug the libraries and your research data management folks for guidance, because in science we have many different file formats and very often we have equipment and software that generates these niche proprietary file formats that are not always useful if we don't have that piece of software, right? So, generally, consider using common formats. Text files are great. Office documents will always be great.
Common image formats are also great, because those are going to be useful 10 years from now, whereas kind of a more unique file type may be more difficult to use. And the last word of fare is reusable. This could probably be an hour in and of itself, we're not going to spend that much time doing that, but talking about the reuse of your work, making it easy to export, making it easy to download, making it consistent, and making it so that if I wanted to query across all of my team's notebooks, pull the result from this sample into one spreadsheet, that's easy to do.
And there are a lot of ways to use lab archives that don't support fair practices. So, we're going to talk about some of the tools that you can use to improve the quality of your research and also support those data management and sharing plans. So, something I recommend to most groups is putting together notebook policies.
So, this is a very rough draft of what that could look like, a very basic one. One is defining the purpose of each notebook. This is going to sound pretty straightforward, but telling people why does this notebook exist, what is supposed to be in it, is very helpful.
Because if you have someone who is putting the wrong data in the wrong notebook, you're not going to be able to find it in the future, because they weren't putting it in the right place. Generally, we do recommend documenting the permissions, as in who had access, how long did they have access, when were they added, when were they removed, what roles did they have. And then, some consistency.
So, you see at the top here, I have page and folder names. So, you can name your pages whatever you want. You can name your folders whatever you want.
You can name your files whatever you want. But if that's not descriptive, if it doesn't tell me what's in that file, it's going to be harder to find the work. So, I usually recommend establishing common naming formats that the whole team uses.
Because if everyone's consistent, they have a trial number, a sample number, start dates, researcher's initials, whatever else, it's going to help me find that data 10 years from now. Because I can look back and find exactly what's going on. Now, something that's worth pointing out, a data management and sharing plan is not a static plan.
This is something that changes. As you get new software, as you get new equipment, as you develop new processes, you have new people come on board, this changes. And it should, right? As new tech moves in, as, you know, new applications and workflows join your group, great.
That's excellent. Document what changes. Document what your process is.
And this will save you time. Because you're having to spend less time having to reinvent the wheel every single experiment. But also, when you have new employees join your team, add them to your notebook policies.
Add them to your data management and sharing plans. So, they have the guidance that is set up for your team. So, on this page here, I'm just showing some examples of some things that I would recommend documenting and how you could document them.
Further down here, this is actually the most recent NIH data management and sharing plan template. Off the top of my head, I believe, Columbia, do you all use DMV tool? Oh, I do. Okay.
There's also a tool called DMV tool that Columbia might use. But if not, talk to your Columbia folks. I'm sure they'd be happy to get you some templates, some language, some recommendations for data management and sharing plans.
You know, because sometimes these questions can be a little hard to piece together. So, some sample language is helpful. And for a lot of folks, I recommend DMV tool because they actually have specific... Oh, perfect.
Thank you, Jessica. And yes, they do have reference to DMV. Awesome.
I figured so, but I don't want to double check. So, I pulled up a document. This is for Columbia's research page that has all about writing a data management and sharing plan.
You'll notice in this, it also is talking about the entire process of your work, archiving, preservation of your work, finding repositories, publishing, even things like software, hardware, budgeting, personnel. There's a lot that goes in. And I recommend reaching out to these folks for some guidance, checking the MP tool, checking some of the various funders requirements and suggestions.
So, you're not having to reinvent the wheel, right? There are many, many, many researchers doing exactly what you do who have worked together to develop some of these templates to save us all some time. So, do check that out. Now, let's talk about the main topic for today.
What can I do in LabArchives to support that data management and sharing plan? The first and kind of best recommendation is a template. So, we're going to talk more about page templates, forms and widgets in the next session, but I'll talk about this briefly now as well. This is a pre-formatted page because in science, we are doing the same thing day in and day out.
And having things pre-formatted saves everyone time. The new grad student who just joined, they have instructions, right? So, they know what they need to do. Similarly, you know, if you run the same experiment this week as last week, copy what you did, update what changed, right? So, you can get back to doing your work so you spend less time documenting it.
If I click the word new in any notebook, I can add a new folder. I can add a new page and I can copy. So, I'm actually going to copy my experimental template.
This is going to copy everything. Perhaps this was a prior experiment. Maybe I don't want to include certain entries.
You can deselect those. And we have the page name. So, I mentioned this before.
Anytime you see an option to put in text, spend a few seconds thinking about how you name it. So, maybe my group, we're going to call it experiment 102. The start date of the experiment.
Again, this is one of those recommendations for the research data management folks. You can have a whole conversation about how you format dates so that it has international standardization. But this is the format I'm choosing today.
It's month to year. And we're going to add my initials. HAC.
Click create. Now, in my notebook, I just have a handful of pages. But you can imagine with a notebook that has 10,000 pages, I can quickly scroll through and say, oh, well, this is experiment 102.
I can tell that. This is October 28th, 2026. And it was work done by Hannah.
So, I actually know a lot about this page without even needing to open it up. Now, because this was a copied page, I have all of these preformatted entries ready for me to go. And I would start at the top and work my way down.
Now, this example is really kind of broad. It just has insert purpose here, insert procedure here. But you could absolutely make these more specific to your industry, your process, and what your team does.
So, this is an example more from the wet lab side of the world where I'm planning a cloud experiment in Snapgene. So, I might have notes about the vector, calculations that I'm performing about melting temperatures, materials that I've used, and even perhaps like a vendor supplied protocol where we are expected to actually kind of check off each step as we complete it. So, this can be a really great way to sort of have standard processes that you complete all the time, copy those templates, check off as you complete.
And consistency is definitely the top priority when we talk about data management and sharing. Now, on our page, what else can we do to increase the quality of our work? So, one of the things here is our file names. So, if I click, we're going to come down here to experimental entries, I'm going to add some attachments.
I'm going to browse my computer. I should have some in the downloads folder. And I would recommend having good, consistent, descriptive file names.
These don't have to be an essay. It doesn't have to take you hours to put it in. But things, again, like the date, sample numbers, IDs, that distinguish what is in this file so I don't have to open it every time to go and say, oh, that's that sample or that experiment or this work.
Similarly, the description is really helpful. Now, let's say this was some proprietary file format that came from our equipment. That's fine.
What I should do in the description or maybe in somewhere else in the notebook is describe how do I open that format? Is it a certain software version? Is it a particular piece of equipment? I should describe that. And descriptions are all going to be searched. So, in LabArchives, you can search pretty much everything you can read.
So, that's going to be file names, file descriptions. So, we will say this is part of project ABCD for trial 1234. And I could include software versions and so forth.
And there's our office document. Now, a brand new feature, which I apologize to the Club AT, you might forget if we've enabled this for you all, but it's brand new and should be enabled in the foreseeable future, is this option right here. Extract text or handwriting.
This is brand new. Came out this year. It's enabled at various institutions that request it.
It allows you to pull out text in an image or photo or scan. So, let's say you have handwritten notes. You take a picture with your phone.
Great. Upload it to LabArchives. That extract text will convert that into searchable text.
This is really useful for handwritten notes, observations. You know, for folks who work in the field, a picture is a lot easier sometimes than other methods. So, it will actually extract the text from this image.
It'll take a minute or two. Then you'll see that on the page as well. It will do its best to convert the text into a string of text that you can search.
It might not be perfect 100%, but you can go back and correct it if it got something wrong. Now, as we've added this work to the notebooks, we've added our experimental entries, our attachments, our descriptions, our extract text, all of that is searchable. I can also consider using tags.
So, tags are a way to build out a data dictionary. Terms, labels, identifiers that mean something to you and to your team. So, let's say I have a term I want to add to my notebook for this particular entry.
I can add that. More importantly, I can select from other previously used tags. So, one of my favorite examples of this is a lab, every time they had a failed experiment, they had a failed experiment tag, and they would add that to the entry.
Or maybe I have a tag for attention PI to notify the PI, hey, this is something they need to take a look at. And these tags can be used across the notebook. So, maybe I add an analysis tag here, then in another part of my notebook, I can add that same analysis tag.
This is a way to build a controlled vocabulary. If you talk with our research data management teams, they would recommend building out ontologies and data dictionaries defining what does the word analysis mean? Does it mean you've performed analysis? You want to perform analysis? This is the analyzed data. Defining what that is is going to be useful for you, yourself, your colleagues, and people down the road.
So, having that in the search is useful. Now, we've added a good amount of stuff. So, we've added some rich text entries, some headings, some attachments.
Up here at the top, I can now search. So, we're going to search our current notebook. Everything by Hannah.
Maybe we can add a filter for the date. We'll run this quick search. So, we are going to find all of the entries that match that particular filter.
I picked the wrong Hannah. I apologize. Here we go.
So, here we see everything that Hannah has done the last couple of days. And I can even sort these by newest to oldest to see the most recent changes that Hannah has made. So, here I see that she was working on a widget.
She was working on a form. And I can click on these links to learn more about what was going on there. It's a really great way to kind of keep up to date with what's going on in your team.
A lot of time, the search is used for management within a group as well. So, let's say I'm a PI. I have five minutes free.
I'm going to quickly come in and just search the word attention PI. And every entry that has that tag, I'm going to take a look at it. Or every entry that has, you know, let's say it's a great picture.
I'm doing some imaging. I find a really great picture. I think it actually could be one we could use in a publication.
I can add a tag so that I can find that more easily down the road when I'm looking for, oh, I want that picture we made that was the good one. That can help you to find that more quickly. Again, notebooks over time could have lots of entries, lots of pages.
So, having good consistent search is all part of that. All right. So, I wanted to pause for a second and see if there were any questions on anything so far.
We talked about kind of forms and widgets and other options there. We'll go into more detail in the next session on that, but I wanted to bring that up now as well. All right.
So, just in case folks aren't joining the next session, we talk about widgets and templates. That's another way to have standardization. So, a widget is a calculator.
There are a bunch of built into archives automatically and you can build your own. So, this is an example. So, my group has the sample ID, researcher name, a collection date, and you can pick on a little calendar and you can do some math and have required fields, drop down list, check boxes.
This is all about consistency. When Jim, Jill, John, and Jenny all do their cell counting, they're all using my widget. They all should be doing it correctly, and I know they're not going to have mental math mistakes or forget to include a sample ID because these things are required.
And when I click save, you see our cell counting. The other side of that is thinking about things like forms. So, forms are going to be very similar to a widget.
We'll talk about these more in a second. The difference is that a form actually reports the data back to form center. So, everything you input with a form, I can then query across my notebooks.
Every time this form is used, pull that into a nice, beautiful spreadsheet for me to analyze, and you can easily pull that when you need to and fill in the data there with that form. So, we'll talk about this more in about an hour with the next session. So, a couple of entry types.
What I usually recommend for groups is a lot of these kind of basic headings, rich text entries, and attachments. And headings are a really great way to divide up a page. Since our pages can get quite long, it's a way to kind of call out a different part of an experiment, a different part of a page, especially if you start having to scroll down real quick.
That can be a nice way to call stuff off. All right. So, talked about a couple of entry types.
Let's talk about finalizing a page and permissioning. So, the previous session, we talked about how to add people and invite people. All of that same information applies.
If I go to account manager, this is where I can invite someone to my notebook, where I can remove someone to my notebook. This is going to sound like common sense, but don't invite people who are not authorized to access the data. Sometimes that's not apparent to folks, but just be aware, you know, who can add data, who can edit data, and, again, least privilege is generally how we should operate.
If you don't need access to a certain notebook, if you don't need access to a certain dataset, you should not have access. There are different lab philosophies, different ways of working, and, you know, I always recommend, you know, you don't want it to be inconvenient, where you have restricted access so people can't see things they need, but you don't want people to be able to access projects they're not part of, datasets they shouldn't be aware of, right? Especially if you're doing, like, double blind studies or anything like that, be very cognizant, who can do what, when they can do it, and, you know, why, right? So, we talked about permissions in the previous session, so I won't go into the details on that, but notebook ownership, again, is going to be something you want to keep in mind, no matter what you do, and the user management, you can see who owns this notebook, so, again, supporting good research data management practices, supporting the long term, know who owns the notebooks, if that owner is leaving before their last day, please be sure to transfer ownership to someone else. A department head, a supervisor, your site admins can take ownership if they really need to, you know, someone, so please make sure that's done before that person leaves.
Sometimes the question comes up, well, what happens if I'm leaving and I want to take my data with me when I leave? There's ways to do that, it's actually very straightforward, but we need to make sure that there are data use agreements in place, that legal is all okay with that, departments are cool with that, it's very common that we have to handle folks kind of moving between organizations, it's really pretty a seamless process, realistically. All right, so we're going to come back to our page, we've talked about adding some entries, file names, page names, opportunities for building out that metadata and how that helps with search. Let's talk a little bit about finalizing our work.
Now, this is something that can vary depending on the industry that you're in, the funders that you have, often this is kind of it, right, experiment 102 is done, I'm going to go and make a new page for experiment 103. In this instance, experiment 102, I am still the owner of this notebook, I can go in and edit these entries whenever I want, I can add data, I can edit data, I can revert to earlier versions, whatever. For some use cases, there's a need to sign pages.
Now generally, this is going to be used for FDA compliant folks, for any time where an intellectual property attorney gets involved, this is going to be one of their recommendations. But also any instances where there's patents, if you ever work with corporate partners, if you ever have, you know, maybe your goal as a lab, right, you're developing a technology and your hope is to spin out in your own little biotech. Cool, that's super exciting.
You're probably going to need page signing. But there's also labs that are not required to use page signing that like to use it. So let's talk about how to set that up.
Triple dot menu, if you go to notebook settings, you have this option here. So by default, this option is selected, page signing only, no witness. This means one person signs the page, as soon as it is signed, it's permanently locked.
Nothing can be added, changed, or removed. Often, we recommend page signing with witness. So I'm going to show what this looks like, basically, there are two people involved.
And you can determine under user management, who has the permission to do that with these checkboxes right here. So in my case, both members of my notebook can sign and can witness. Very often that witness person is a more senior person, a supervisor, a more senior graduate student, a lab manager, a quality assurance, quality control person.
It doesn't have to be, sometimes it is just kind of a fellow lab tech. It gives you the ability to have a sort of check and balance, right? I can sign my page, but someone else is going to double check my work, double check my math, right? And well, let's show how we do that. So we have run experiment 102, we've collected our data, we've analyzed our data, we have all these beautiful tags and labels and everything else.
When I'm ready, I click page tools, and I click sign page. Now, please read the yellow warning at the top here. By clicking sign and notify witness, I have reviewed the contents of this page and I'm signing, right? I am locking this down.
Now this comes from the paper notebook days. It is a legal record. That's it.
I'm done. I agree that this is it. We're not going to add anything after this point.
You all are not going to need to fill this out, but I will because my account is not part of Columbia. So one second. When I click sign and notify witness, the page is locked.
So you see the little lock icon. You'll also see a little exclamation icon on the page. And I selected this witness of this person here.
Sometimes you have several. So maybe Hannah Clark's out on vacation. Someone else is filling in.
I can add the other witness. And in general, it is good practice that if you use page signing to sign the page in a timely manner, it doesn't have to be 24 hours, but I would say within a week, maybe two weeks. Because if you go three years down the road, you will not remember that Hannah asked you to sign something.
You will lose track of which page that was. That's why there's a little exclamation icon to let you know, hey, this isn't pending. It's also why I often recommend having a folder for this purpose.
So maybe we have a pending review folder where these pages get moved to. So it's nice and clear to everyone in our group. If it's in that folder, someone needs to witness.
And of course, you can change the witness. You notice there's no option to unsign the page. I can't do anything.
I can't edit. I can't add. It's locked.
Now, I'm going to stay in the same account today. So I preset up a page so we can show this. Typically, someone else would come along to our page and review it.
So that's what you see here. My colleague signed the page and asked me to witness. So you'll notice I still can't edit as the witness.
I can't add anything. If I click witness, I get asked to, again, verify my information, confirm I want this page locked down. But if I read through, there's some mistakes here.
I don't like that this column is blank or something's not quite right. I can click reject and give feedback. This is really helpful because sometimes the person, maybe they didn't name the page properly.
Maybe they didn't add the folder properly. Maybe they didn't do math properly. Something just isn't quite right.
Here's some suggestions for a next step. It's a really great opportunity as a manager to give feedback to newer employees, more junior colleagues, to kind of give them some guidance on how they should be documenting things. Now, I think this page looks good.
Like the word, looks big, everything's named properly, the date and timestamps all check out. I'll click witness. Again, you won't need to do this step.
It's just because I am not part of Columbia. If you do reject the original person, you can go back and make edits. Yes, and we can do that instead.
If I hit witness, it'll show both signatures. It'll be permanently locked. Let's say instead I say, no, Hannah, this looks like a mess.
We can reject immediately. The page is unlocked. I can start adding.
And sometimes the person who rejected it immediately fixes the issue and just goes, this is what needs to be done and resigns, which is fine. It's also worth bringing out permissions for page signing and witnessing is per notebook. Sometimes someone is a witness in one notebook, but not in another.
And sometimes one project or workflow doesn't require page signing at all. So don't worry about it. Generally, for a lot of labs, I say don't worry about page signing and witnessing until an attorney tells you to worry about it, in which case you need it.
So it's something to think about. But you'll see here our page was rejected. I can come in here and edit and add things.
And when I'm ready, I can sign it again and lock it back down. So it gives that option to go back and forth. Now, we've talked a bit about page signing and witnessing.
I wanted to pause, see if there were any questions on the call or in the room. All right. So last couple things we want to talk about is reuse of data, right? So this is something that's very, probably the most important thing when we talk about FAIR data is reuse.
So it's important to think with your team, how do you intend to reuse your work? This has been a very popular topic as more and more folks are looking into AI options, right? If I want the data to be analyzed by an AI tool, I should add it in a way that an AI tool could use it. Or if I want my data to be exportable to a certain format, I should make sure that it's exportable to that format. On the LabArchive side, we are all about reproducibility.
So if I upload a PNG file that is 84.5 kilobytes and I download it, guess what? I get a PNG image with the exact same file name, with the exact same file type and size. That's important. And the reason I bring that up\
So as an example, NIH publishes excellent templates that are available on the various NIH websites for specific information, and these actually need to be filed with NIH. But I would also volunteer your excellent libraries on campus and your research data management folks who can give you some guidance on good data management practices, good practices in your particular area of research, not just, you know, broad, but more kind of your discipline, but also how the data management and sharing plan affects your whole life cycle of your data. So from coming up with an idea to documenting, collecting your data, to analyzing it, to preparing figures and preparing your publication, all the way until looking for repositories, publishing of data, public access of your work.
So kind of that full research life cycle. Now you'll often hear in the industry an acronym of FAIR, findable, accessible, interoperable, and reusable. So what supports findable data? Search is the simple answer, but in LabArchives we have great search.
You can type in a term, you can click the drop down and see our advanced search tools. So you can search by notebook, by user, by guest, by entry type, by page title, and look for the data that you need. Now the problem is your search is only as good as your data.
So if your team is just uploading files that are called untitled one, or using really inconsistent page names, really inconsistent folder names, or you have no consistency at all, right? Day-to-day your documentation is going to be different. It's harder to search for your work. Now there are plenty of ways to help improve search and we're going to talk about it.
Accessible is a pretty involved process talking about both access, being able to get to your notebooks, even if you're working remotely, but also permissioning. We'll talk about that as well. Interoperable is all about being able to use your data in the long term, right? Sure, you might develop this code, this script, this work that works great on a specific version of Java, a specific version of PHP, a specific version of Chrome on a Windows computer, but you should also make sure it works on other operating systems, it works in other systems, and will be usable for the long term.
Again, this is one of those areas where I would plug the libraries and your research data management folks for guidance, because in science we have many different file formats and very often we have equipment and software that generates these niche proprietary file formats that are not always useful if we don't have that piece of software, right? So, generally, consider using common formats. Text files are great. Office documents will always be great.
Common image formats are also great, because those are going to be useful 10 years from now, whereas kind of a more unique file type may be more difficult to use. And the last word of fare is reusable. This could probably be an hour in and of itself, we're not going to spend that much time doing that, but talking about the reuse of your work, making it easy to export, making it easy to download, making it consistent, and making it so that if I wanted to query across all of my team's notebooks, pull the result from this sample into one spreadsheet, that's easy to do.
And there are a lot of ways to use lab archives that don't support fair practices. So, we're going to talk about some of the tools that you can use to improve the quality of your research and also support those data management and sharing plans. So, something I recommend to most groups is putting together notebook policies.
So, this is a very rough draft of what that could look like, a very basic one. One is defining the purpose of each notebook. This is going to sound pretty straightforward, but telling people why does this notebook exist, what is supposed to be in it, is very helpful.
Because if you have someone who is putting the wrong data in the wrong notebook, you're not going to be able to find it in the future, because they weren't putting it in the right place. Generally, we do recommend documenting the permissions, as in who had access, how long did they have access, when were they added, when were they removed, what roles did they have. And then, some consistency.
So, you see at the top here, I have page and folder names. So, you can name your pages whatever you want. You can name your folders whatever you want.
You can name your files whatever you want. But if that's not descriptive, if it doesn't tell me what's in that file, it's going to be harder to find the work. So, I usually recommend establishing common naming formats that the whole team uses.
Because if everyone's consistent, they have a trial number, a sample number, start dates, researcher's initials, whatever else, it's going to help me find that data 10 years from now. Because I can look back and find exactly what's going on. Now, something that's worth pointing out, a data management and sharing plan is not a static plan.
This is something that changes. As you get new software, as you get new equipment, as you develop new processes, you have new people come on board, this changes. And it should, right? As new tech moves in, as, you know, new applications and workflows join your group, great.
That's excellent. Document what changes. Document what your process is.
And this will save you time. Because you're having to spend less time having to reinvent the wheel every single experiment. But also, when you have new employees join your team, add them to your notebook policies.
Add them to your data management and sharing plans. So, they have the guidance that is set up for your team. So, on this page here, I'm just showing some examples of some things that I would recommend documenting and how you could document them.
Further down here, this is actually the most recent NIH data management and sharing plan template. Off the top of my head, I believe, Columbia, do you all use DMV tool? Oh, I do. Okay.
There's also a tool called DMV tool that Columbia might use. But if not, talk to your Columbia folks. I'm sure they'd be happy to get you some templates, some language, some recommendations for data management and sharing plans.
You know, because sometimes these questions can be a little hard to piece together. So, some sample language is helpful. And for a lot of folks, I recommend DMV tool because they actually have specific... Oh, perfect.
Thank you, Jessica. And yes, they do have reference to DMV. Awesome.
I figured so, but I don't want to double check. So, I pulled up a document. This is for Columbia's research page that has all about writing a data management and sharing plan.
You'll notice in this, it also is talking about the entire process of your work, archiving, preservation of your work, finding repositories, publishing, even things like software, hardware, budgeting, personnel. There's a lot that goes in. And I recommend reaching out to these folks for some guidance, checking the MP tool, checking some of the various funders requirements and suggestions.
So, you're not having to reinvent the wheel, right? There are many, many, many researchers doing exactly what you do who have worked together to develop some of these templates to save us all some time. So, do check that out. Now, let's talk about the main topic for today.
What can I do in LabArchives to support that data management and sharing plan? The first and kind of best recommendation is a template. So, we're going to talk more about page templates, forms and widgets in the next session, but I'll talk about this briefly now as well. This is a pre-formatted page because in science, we are doing the same thing day in and day out.
And having things pre-formatted saves everyone time. The new grad student who just joined, they have instructions, right? So, they know what they need to do. Similarly, you know, if you run the same experiment this week as last week, copy what you did, update what changed, right? So, you can get back to doing your work so you spend less time documenting it.
If I click the word new in any notebook, I can add a new folder. I can add a new page and I can copy. So, I'm actually going to copy my experimental template.
This is going to copy everything. Perhaps this was a prior experiment. Maybe I don't want to include certain entries.
You can deselect those. And we have the page name. So, I mentioned this before.
Anytime you see an option to put in text, spend a few seconds thinking about how you name it. So, maybe my group, we're going to call it experiment 102. The start date of the experiment.
Again, this is one of those recommendations for the research data management folks. You can have a whole conversation about how you format dates so that it has international standardization. But this is the format I'm choosing today.
It's month to year. And we're going to add my initials. HAC.
Click create. Now, in my notebook, I just have a handful of pages. But you can imagine with a notebook that has 10,000 pages, I can quickly scroll through and say, oh, well, this is experiment 102.
I can tell that. This is October 28th, 2026. And it was work done by Hannah.
So, I actually know a lot about this page without even needing to open it up. Now, because this was a copied page, I have all of these preformatted entries ready for me to go. And I would start at the top and work my way down.
Now, this example is really kind of broad. It just has insert purpose here, insert procedure here. But you could absolutely make these more specific to your industry, your process, and what your team does.
So, this is an example more from the wet lab side of the world where I'm planning a cloud experiment in Snapgene. So, I might have notes about the vector, calculations that I'm performing about melting temperatures, materials that I've used, and even perhaps like a vendor supplied protocol where we are expected to actually kind of check off each step as we complete it. So, this can be a really great way to sort of have standard processes that you complete all the time, copy those templates, check off as you complete.
And consistency is definitely the top priority when we talk about data management and sharing. Now, on our page, what else can we do to increase the quality of our work? So, one of the things here is our file names. So, if I click, we're going to come down here to experimental entries, I'm going to add some attachments.
I'm going to browse my computer. I should have some in the downloads folder. And I would recommend having good, consistent, descriptive file names.
These don't have to be an essay. It doesn't have to take you hours to put it in. But things, again, like the date, sample numbers, IDs, that distinguish what is in this file so I don't have to open it every time to go and say, oh, that's that sample or that experiment or this work.
Similarly, the description is really helpful. Now, let's say this was some proprietary file format that came from our equipment. That's fine.
What I should do in the description or maybe in somewhere else in the notebook is describe how do I open that format? Is it a certain software version? Is it a particular piece of equipment? I should describe that. And descriptions are all going to be searched. So, in LabArchives, you can search pretty much everything you can read.
So, that's going to be file names, file descriptions. So, we will say this is part of project ABCD for trial 1234. And I could include software versions and so forth.
And there's our office document. Now, a brand new feature, which I apologize to the Club AT, you might forget if we've enabled this for you all, but it's brand new and should be enabled in the foreseeable future, is this option right here. Extract text or handwriting.
This is brand new. Came out this year. It's enabled at various institutions that request it.
It allows you to pull out text in an image or photo or scan. So, let's say you have handwritten notes. You take a picture with your phone.
Great. Upload it to LabArchives. That extract text will convert that into searchable text.
This is really useful for handwritten notes, observations. You know, for folks who work in the field, a picture is a lot easier sometimes than other methods. So, it will actually extract the text from this image.
It'll take a minute or two. Then you'll see that on the page as well. It will do its best to convert the text into a string of text that you can search.
It might not be perfect 100%, but you can go back and correct it if it got something wrong. Now, as we've added this work to the notebooks, we've added our experimental entries, our attachments, our descriptions, our extract text, all of that is searchable. I can also consider using tags.
So, tags are a way to build out a data dictionary. Terms, labels, identifiers that mean something to you and to your team. So, let's say I have a term I want to add to my notebook for this particular entry.
I can add that. More importantly, I can select from other previously used tags. So, one of my favorite examples of this is a lab, every time they had a failed experiment, they had a failed experiment tag, and they would add that to the entry.
Or maybe I have a tag for attention PI to notify the PI, hey, this is something they need to take a look at. And these tags can be used across the notebook. So, maybe I add an analysis tag here, then in another part of my notebook, I can add that same analysis tag.
This is a way to build a controlled vocabulary. If you talk with our research data management teams, they would recommend building out ontologies and data dictionaries defining what does the word analysis mean? Does it mean you've performed analysis? You want to perform analysis? This is the analyzed data. Defining what that is is going to be useful for you, yourself, your colleagues, and people down the road.
So, having that in the search is useful. Now, we've added a good amount of stuff. So, we've added some rich text entries, some headings, some attachments.
Up here at the top, I can now search. So, we're going to search our current notebook. Everything by Hannah.
Maybe we can add a filter for the date. We'll run this quick search. So, we are going to find all of the entries that match that particular filter.
I picked the wrong Hannah. I apologize. Here we go.
So, here we see everything that Hannah has done the last couple of days. And I can even sort these by newest to oldest to see the most recent changes that Hannah has made. So, here I see that she was working on a widget.
She was working on a form. And I can click on these links to learn more about what was going on there. It's a really great way to kind of keep up to date with what's going on in your team.
A lot of time, the search is used for management within a group as well. So, let's say I'm a PI. I have five minutes free.
I'm going to quickly come in and just search the word attention PI. And every entry that has that tag, I'm going to take a look at it. Or every entry that has, you know, let's say it's a great picture.
I'm doing some imaging. I find a really great picture. I think it actually could be one we could use in a publication.
I can add a tag so that I can find that more easily down the road when I'm looking for, oh, I want that picture we made that was the good one. That can help you to find that more quickly. Again, notebooks over time could have lots of entries, lots of pages.
So, having good consistent search is all part of that. All right. So, I wanted to pause for a second and see if there were any questions on anything so far.
We talked about kind of forms and widgets and other options there. We'll go into more detail in the next session on that, but I wanted to bring that up now as well. All right.
So, just in case folks aren't joining the next session, we talk about widgets and templates. That's another way to have standardization. So, a widget is a calculator.
There are a bunch of built into archives automatically and you can build your own. So, this is an example. So, my group has the sample ID, researcher name, a collection date, and you can pick on a little calendar and you can do some math and have required fields, drop down list, check boxes.
This is all about consistency. When Jim, Jill, John, and Jenny all do their cell counting, they're all using my widget. They all should be doing it correctly, and I know they're not going to have mental math mistakes or forget to include a sample ID because these things are required.
And when I click save, you see our cell counting. The other side of that is thinking about things like forms. So, forms are going to be very similar to a widget.
We'll talk about these more in a second. The difference is that a form actually reports the data back to form center. So, everything you input with a form, I can then query across my notebooks.
Every time this form is used, pull that into a nice, beautiful spreadsheet for me to analyze, and you can easily pull that when you need to and fill in the data there with that form. So, we'll talk about this more in about an hour with the next session. So, a couple of entry types.
What I usually recommend for groups is a lot of these kind of basic headings, rich text entries, and attachments. And headings are a really great way to divide up a page. Since our pages can get quite long, it's a way to kind of call out a different part of an experiment, a different part of a page, especially if you start having to scroll down real quick.
That can be a nice way to call stuff off. All right. So, talked about a couple of entry types.
Let's talk about finalizing a page and permissioning. So, the previous session, we talked about how to add people and invite people. All of that same information applies.
If I go to account manager, this is where I can invite someone to my notebook, where I can remove someone to my notebook. This is going to sound like common sense, but don't invite people who are not authorized to access the data. Sometimes that's not apparent to folks, but just be aware, you know, who can add data, who can edit data, and, again, least privilege is generally how we should operate.
If you don't need access to a certain notebook, if you don't need access to a certain dataset, you should not have access. There are different lab philosophies, different ways of working, and, you know, I always recommend, you know, you don't want it to be inconvenient, where you have restricted access so people can't see things they need, but you don't want people to be able to access projects they're not part of, datasets they shouldn't be aware of, right? Especially if you're doing, like, double blind studies or anything like that, be very cognizant, who can do what, when they can do it, and, you know, why, right? So, we talked about permissions in the previous session, so I won't go into the details on that, but notebook ownership, again, is going to be something you want to keep in mind, no matter what you do, and the user management, you can see who owns this notebook, so, again, supporting good research data management practices, supporting the long term, know who owns the notebooks, if that owner is leaving before their last day, please be sure to transfer ownership to someone else. A department head, a supervisor, your site admins can take ownership if they really need to, you know, someone, so please make sure that's done before that person leaves.
Sometimes the question comes up, well, what happens if I'm leaving and I want to take my data with me when I leave? There's ways to do that, it's actually very straightforward, but we need to make sure that there are data use agreements in place, that legal is all okay with that, departments are cool with that, it's very common that we have to handle folks kind of moving between organizations, it's really pretty a seamless process, realistically. All right, so we're going to come back to our page, we've talked about adding some entries, file names, page names, opportunities for building out that metadata and how that helps with search. Let's talk a little bit about finalizing our work.
Now, this is something that can vary depending on the industry that you're in, the funders that you have, often this is kind of it, right, experiment 102 is done, I'm going to go and make a new page for experiment 103. In this instance, experiment 102, I am still the owner of this notebook, I can go in and edit these entries whenever I want, I can add data, I can edit data, I can revert to earlier versions, whatever. For some use cases, there's a need to sign pages.
Now generally, this is going to be used for FDA compliant folks, for any time where an intellectual property attorney gets involved, this is going to be one of their recommendations. But also any instances where there's patents, if you ever work with corporate partners, if you ever have, you know, maybe your goal as a lab, right, you're developing a technology and your hope is to spin out in your own little biotech. Cool, that's super exciting.
You're probably going to need page signing. But there's also labs that are not required to use page signing that like to use it. So let's talk about how to set that up.
Triple dot menu, if you go to notebook settings, you have this option here. So by default, this option is selected, page signing only, no witness. This means one person signs the page, as soon as it is signed, it's permanently locked.
Nothing can be added, changed, or removed. Often, we recommend page signing with witness. So I'm going to show what this looks like, basically, there are two people involved.
And you can determine under user management, who has the permission to do that with these checkboxes right here. So in my case, both members of my notebook can sign and can witness. Very often that witness person is a more senior person, a supervisor, a more senior graduate student, a lab manager, a quality assurance, quality control person.
It doesn't have to be, sometimes it is just kind of a fellow lab tech. It gives you the ability to have a sort of check and balance, right? I can sign my page, but someone else is going to double check my work, double check my math, right? And well, let's show how we do that. So we have run experiment 102, we've collected our data, we've analyzed our data, we have all these beautiful tags and labels and everything else.
When I'm ready, I click page tools, and I click sign page. Now, please read the yellow warning at the top here. By clicking sign and notify witness, I have reviewed the contents of this page and I'm signing, right? I am locking this down.
Now this comes from the paper notebook days. It is a legal record. That's it.
I'm done. I agree that this is it. We're not going to add anything after this point.
You all are not going to need to fill this out, but I will because my account is not part of Columbia. So one second. When I click sign and notify witness, the page is locked.
So you see the little lock icon. You'll also see a little exclamation icon on the page. And I selected this witness of this person here.
Sometimes you have several. So maybe Hannah Clark's out on vacation. Someone else is filling in.
I can add the other witness. And in general, it is good practice that if you use page signing to sign the page in a timely manner, it doesn't have to be 24 hours, but I would say within a week, maybe two weeks. Because if you go three years down the road, you will not remember that Hannah asked you to sign something.
You will lose track of which page that was. That's why there's a little exclamation icon to let you know, hey, this isn't pending. It's also why I often recommend having a folder for this purpose.
So maybe we have a pending review folder where these pages get moved to. So it's nice and clear to everyone in our group. If it's in that folder, someone needs to witness.
And of course, you can change the witness. You notice there's no option to unsign the page. I can't do anything.
I can't edit. I can't add. It's locked.
Now, I'm going to stay in the same account today. So I preset up a page so we can show this. Typically, someone else would come along to our page and review it.
So that's what you see here. My colleague signed the page and asked me to witness. So you'll notice I still can't edit as the witness.
I can't add anything. If I click witness, I get asked to, again, verify my information, confirm I want this page locked down. But if I read through, there's some mistakes here.
I don't like that this column is blank or something's not quite right. I can click reject and give feedback. This is really helpful because sometimes the person, maybe they didn't name the page properly.
Maybe they didn't add the folder properly. Maybe they didn't do math properly. Something just isn't quite right.
Here's some suggestions for a next step. It's a really great opportunity as a manager to give feedback to newer employees, more junior colleagues, to kind of give them some guidance on how they should be documenting things. Now, I think this page looks good.
Like the word, looks big, everything's named properly, the date and timestamps all check out. I'll click witness. Again, you won't need to do this step.
It's just because I am not part of Columbia. If you do reject the original person, you can go back and make edits. Yes, and we can do that instead.
If I hit witness, it'll show both signatures. It'll be permanently locked. Let's say instead I say, no, Hannah, this looks like a mess.
We can reject immediately. The page is unlocked. I can start adding.
And sometimes the person who rejected it immediately fixes the issue and just goes, this is what needs to be done and resigns, which is fine. It's also worth bringing out permissions for page signing and witnessing is per notebook. Sometimes someone is a witness in one notebook, but not in another.
And sometimes one project or workflow doesn't require page signing at all. So don't worry about it. Generally, for a lot of labs, I say don't worry about page signing and witnessing until an attorney tells you to worry about it, in which case you need it.
So it's something to think about. But you'll see here our page was rejected. I can come in here and edit and add things.
And when I'm ready, I can sign it again and lock it back down. So it gives that option to go back and forth. Now, we've talked a bit about page signing and witnessing.
I wanted to pause, see if there were any questions on the call or in the room. All right. So last couple things we want to talk about is reuse of data, right? So this is something that's very, probably the most important thing when we talk about FAIR data is reuse.
So it's important to think with your team, how do you intend to reuse your work? This has been a very popular topic as more and more folks are looking into AI options, right? If I want the data to be analyzed by an AI tool, I should add it in a way that an AI tool could use it. Or if I want my data to be exportable to a certain format, I should make sure that it's exportable to that format. On the LabArchive side, we are all about reproducibility.
So if I upload a PNG file that is 84.5 kilobytes and I download it, guess what? I get a PNG image with the exact same file name, with the exact same file type and size. That's important. And the reason I bring that up\