What is your check list to follow before starting research?

There isn't one common checklist. Though many elements may overlap, each project requires its own checklist. Research process is also enormously different between academia and industry, an individual plodding from A to Z being quite common in the former, especially during Ph.D. and often during post-doc as well, while overlapping team-work is the norm in the latter. Starting with a broad-brush breakdown of basic immunology research into either mouse model or in vitro human cell studies,

Mouse model studies

Carefully research pertinent literature and draft an Animal Study Proposal (ASP). Submit ASP to the Institutional Animal Care and Use Committee (IACUC) and wait for their approval. Experiments involving animal models require prior approval by IACUCs. Typically, IACUCs meet once a month so already we see how planning is integral to basic biomedical research, especially if it involves animal models. The IACUC process ensures ethical animal use and is mandated by law.

ASPs need to detail how many animals are needed for a year, how many experiments, how many animals per year, age- and gender-matched or not, plus clear scientific rationale for each choice. ASPs also need to account for situations involving unrelieved pain and distress. Would any animal be exposed to such? If yes, then need to explain why this is scientifically necessary and also need to scientifically justify the numbers of such animals. ASPs typically undergo annual renewal, at which point changes in experiment designs, numbers, especially for increases in those likely to be subjected to unrelieved pain and distress need to be rigorously scientifically justified.

Special Knockout mouse, Genetically modified mouse require more extensive time outlays in creation and breeding. If procured commercially, need to factor cost and availability as well.

Human cell studies

Carefully research pertinent literature and draft a human research study protocol and submit to the Institutional review board (IRB). Typically, if the human cells are just blood cells, i.e., requiring collection of blood samples, then the process is considered minimal risk and merits expedited review (http://www.hhs.gov/ohrp/policy/e...). The IRB review process ensures that research involving human subjects is conducted ethically. Again, process requires advance planning since research can only proceed after IRB approval.

Questions can then be broken down into:

Sufficient supply of animals/human cell needed to do the entire study or not? If not, what's the plan? Proceed or wait?

How many experiments? Per week? Per month? Etc.

How many animals/human cell vials needed/experiment? Gender- and age-matched.

Need specialized media bottles or supplements or not? If yes, are they readily available or tend to be on back-order? If the latter, how long is the back-order? Also, if the latter, need to stockpile such materials prior to starting such an experiment series. In that case, how long such reagents are good for also becomes a critical issue. Some reagents may be good for a year or more, others only for a few months. If the latter, then need to decide if entire experimental series could be done with one stockpile or not. If not, then multiple lots of a specific reagent would be needed and the need to control for this variable needs to be incorporated into the experiment design.

Need Fetal Calf Serum (FCS*) as growth supplement or not? If yes, then need to order enough bottles of one lot of FCS from one particular vendor for an entire experimental series. Typically labs do or should screen for various FCS lots using their most common lab assays as the readouts and choose to purchase one FCS lot necessary to sustain their lab activities for several years. There is enormous lot-to-lot variation in FCS so biomedical experiments using it have to control for this.

Antibodies, assay kits, enzymes, recombinant proteins, other reagents, lab consumables: Need to prepare a checklist of the foreseeable reagents needed for the planned experiments and check their availability from vendors. Experiments should commence only once all the necessary reagents and equipment are available to hand. For e.g., a particular experimental series might require unusual, specialized lab consumables such as moulded 96- or 48-well transwells that may need to be ordered ahead of time and stockpiled for an entire experimental series or may need a standing order delivery of specific number of units periodically.

Protocols and Standard operating procedure (SOPs): Each slated experimental procedure should be clearly and succinctly written down, and shared with all team members who would be performing the experiments. In industry, SOPs may often need to go through a formal review process as well. Pilot experiments involving all team members are very useful to work out the kinks in new protocols and help minimize 'loss in translation' ahead of primetime. Once an experiment series is underway, it's also helpful to have a shared calendar charting all the steps. Different experiments can be color-coded. Shared calendars help work out schedules ahead of time, which in turn helps outline if staff need to come in on week-ends or holidays or not. Shared calendars also help distribute tasks among team members, and generally help keep the experimental pipeline running smoothly. Again, to minimize loss in translation, experiments should use standardized experiment templates, designed ahead of time as much as possible. In research teams therefore, daily conversations and discussions are the norm and necessity.

Data collection, storage and analysis, ontologies, etc.: If some repetitive procedures require complicated calculations or estimations, excel macros and the like, these should be written ahead of time and saved in a common folder accessible to all team members.  

Electronic notebooks are more optimal compared to paper for data capture. Data can be uploaded to a central server as recorded with no scope for post-experiment modifications, only additions. Minimizes scope for fraud, data selection and other unscrupulous or dubious research practices. Team members should use common data analysis macros or cheat-sheets to ensure uniform and comparable data analysis.

Basic theme? Research teams creating and adhering to a common experimental language, i.e., protocols, calendars, templates, calculations, data analysis approaches and ontologies, helps minimize experimental errors, misunderstandings and miscommunications. Each experiment series reveals scope for improvement such as greater granularity of detail required for successful reproducible experimentation. Electronic notebooks help here as well since date- and time-stamped notes can be recorded in real-time and saved at a central location, enabling ease of retrieval at future dates for post-study critique to tweak and improve the research process.

Some aspects explained in greater detail here: Tirumalai Kamala's answer to What are your hearty suggestions to the students new to research?

* FCS is used inter-changeably with FBS (Fetal Bovine Serum).

https://www.quora.com/What-is-your-check-list-to-follow-before-starting-research/answer/Tirumalai-Kamala