“What I wish I had known before starting my PhD”

Abstract As a rather recent PhD graduate and still an “early career researcher”, the author wondered what to write about that would be interesting for a young scientist. The answer came while overhearing students in the break room stating, “I wish I had known all that before starting my PhD that would have made everything easier!” – An experience many researchers are very familiar with. From simple tricks for laboratory work to choosing the right software or planning the next career steps, this was a reoccurring theme during the career of the author, who will try to give a short personal overview for young researchers, especially in the analytics and/or natural products field. These topics and lists represent a personal opinion and are neither meant to be all‐encompassing nor of course might differ from the experiences of other researchers.


"BE AWARE OF NEW TECHNOLOGIES AND SOFTWARE"
New technologies and method development lead to consistent improvement and implementation of new laboratory procedures.The analytical science field is thereby especially driven by the development of novel and improved equipment.In the field of, for example, chromatography and mass spectrometry (MS), large advances in the last decades, have not only implemented new techniques such as 3D high-performance liquid chromatography (HPLC), modern ultraperformance liquid chromatography systems or improved stationary phase materials to increase the separation efficiency and throughput of liquid chromatography but likewise saw a stark increase in the capability of mass spectrometers themselves.From MS-based imaging techniques 5 to utilizing ion mobility 6 or the development of magnetic resonance mass spectrometers, 7 new technologies consistently improve what can be achieved by (LC-) MS equipment.Recent technologies further improved measurement speeds, by replacing the LC-separation through, for example, an automated solid-phase extraction (SPE) step (SPE-MS) 8 or acoustic droplet ejection 9 to utilizing acoustic mist ionization MS 10 for "ultra-high throughput" MS. 11 The increase of measurement speeds and the corresponding amount of data however also highlights the need for more streamlined and automated analytic workflows.
While young researchers will rarely have the chance to choose which instrument to acquire, it is always wise to stay up to date with recent developments and trends.3][14] While most researchers often start using vendor-specific software, the question arises on how to deal with larger or very specific datasets, for example, in the field of proteomics, metabolomics or MS imaging (MSI).
Luckily, there is not only a wide variety of commercial and free software available but also a growing analytics/bioinformatics community driving further improvements.The diversity and limitations however can make it hard, even for experienced researchers, to know which tools to choose or test out.Most analytical laboratories will already have their favourites and experienced colleagues to ask, but if you are vendor-independent software however is also independent of specific raw file formats and works with universal MS data formats such as mzML (or imzML for MSI), to which all raw data can be converted with free tools such as MS convert (Proteowizard Toolbox, Table 1).Have a look through the cited reviews and software to find suitable software to aid and speed up your data analysis.While the right choice of course depends on your research goals, a short exemplary selection of project-specific software can be found in Tables 1-4.Furthermore, the following advice was always useful to the author:

2.2
How old is the software, is it still being maintained/updated (e.g. on GitHub)?
Especially with non-commercial software, there is no guarantee that it is further developed, or any support is available.As it can be very frustrating choosing software for your analysis workflow, just to discover that you cannot fix small bugs, always check if there is an active community using the software (and citing it) and if there is support available if you get stuck.

Does the software have a graphical user interface or is it based on scripts?
Even though most commercial scientific software has more-or-less intuitive graphical user interfaces (GUIs), to ease the use of the software, many very powerful non-commercial tools are often based on using scripts written in coding languages, for example, R or Python.For users with no bioinformatics background, using software with a good GUI can reduce a lot of complexity at the beginning.However, from personal experience, the author can only recommend trying to get a basic understanding of these coding languages and trying out freely available scripts published in repositories such as GitHub.A lot of them are driven and consistently optimized by a large scientific community and offer the opportunity to tailor a workflow very specifically and ask about it.
Especially when students are just starting out, it is very helpful to have a few standard workflows and methods to follow.The more you

"Money is always tight -But there are tricks to safe it"
Even though the founding of research groups can differ massively, even within departments, money is always an issue.Everything, from scientific software and access to publications to single-use equipment, is the limited resources on.There are however some good ways to save money and resources that might not be obvious to young researchers.
As described above, there is a great diversity of free software and databases available, which are often able to replace expensive licensed variants.While, for example, in the metabolomics fields, software like MassHunter (Agilent) or MetaboScape (Bruker), are often the first choices depending on used instrumentation, freeware such as MS DIAL 31 or MzMine 33,34 can likewise accomplish many of the offered functions of licensed software.Similarly, one has to evaluate if money should be allocated to purchase spectral metabolite databases, for example, Metlin Gen2 38 or the NIST spectral database (Table 2) or if freely available and community-driven databases such as GNPS 35 or Massbank 36 are already sufficient for one's approach.
Due to the increase in sample sizes and implementation of HTP workflows, the amount of single-use plastic has recently become a focus for financial and ecological awareness also in academic settings. 63,64There can be of course good arguments for single-use equipment, for example, in GMP or medical settings, but in many situations, the reuse of disposable equipment is absolutely possible, even though discouraged by the vendors for obvious reasons.The author can only urge young researchers to think about their workflow, and used equipment and search for, or develop reuse protocols when possible and reasonable.There are already multiple protocols published to regenerate and reuse "disposable" equipment starting from cleaning reaction tubes to DNA-purification columns, 65,66 protein filter units 67,68 up to SPE filter tips, columns or disks. 69The reuse of equipment of course carries the risk of sample contamination through

A few last words
The author of this text, like most (former) PhD students, has experienced great and exciting, as well as very frustrating days or weeks where nothing worked the way, it was supposed to.While the latter can be especially demotivating, learning from mistakes and testing different approaches is as much a part of the learning process as finally getting those publishable results.This however does not mean that you cannot learn from the experiences of others and avoid the same frustration or mistakes.The author hopes that the advice given here will motivate young researchers to make use of state-of-the-art techniques and software, safe resources where possible and maybe even find new approaches to make your PhD life a little easier and more exciting.

2.1 Vendor-specific or external software?
comes with software to view your spectra and perform basic operations like creating an extracted ion chromatogram.They often offer additional (expensive) software for specific fields, for example, protein analysis or metabolomics.Positive hereby is an often-well-designed user interface and available support and training.On the other hand, most of these are specific for data recorded with one instrument, thus cannot analyze data stemming from "foreign" instruments, which can be critical if the research groups own diverse instrumentation.Most

.5 Reach out to other researchers/collaborate
Authors' selection of generally useful software that young researchers might benefit from Editorial doi.org/10.1002/ansa.202200044you want to end up, look up these positions and what requirements they have early and try to learn some of them when possible.Especially if you want to transfer into the industry, make sure your skills are needed there or even with a large publication output you will have limited chances to get the position you want.
TA B L E 4 much essential for any future leadership roles, in which you have to supervise students or employees.In the same regard, invest in project management and organizational skills to keep your work on track.In general, obtain useful additional skills, even if they are not essential for your current research.No matter if you want to stay in academia or find a position in the industry, think about your CV and the skills you have and might need in the future.If you have a basic idea of where