Neuroinformatics a subset of Bioinformatics?

Question

Proposal: Neuroinformatics

I've only worked with Bioinformatics peripherally, and a few years ago, but how is Neuroinformatics NOT a subset of it?

Answer 1

Technically speaking, neuro is a subset of bio and this would also be true as per wikipedia's definition of bioinformatics. But practically speaking, bioinformatics has primarily targeted genomics and proteomics. Even in the definition and description page on wikipedia, the only place there is a mention of anything neuro is in the definition only "neural computation".

Neuroinformatics serves to focus on the intersection between neuroscience and informatics. There are points of overlap, but the practical reality is that these are currently quite separate domains of expertise. Even as evidenced by the current trends in questions.

It looks like there is yet another place - biostar exchange site - a presumably even broader group. And I don't know at what point in time or if these should all merge.

Answer 2

I think it is important to remember that informatics in biomedicine, as dnkennedy articulates, spans many domains, and I would add that the scientific discipline of Biomedical Informatics encompasses the domains ranging from biology and clinical care to public health. Although Wikipedia inaccurately calls it Health Informatics, also see AMIA.

What makes neuroinformatics different is the system under study, which spans from the world of genomics and bioinformatics, to encompass brain functioning at the neuronal, neural circuit, cortical field, and cognitive levels. In my mind, and in the literature, bioinformatics does not address issues beyond the cellular/sub-cellular level, yet biology includes so much more.

Another aspect of informatics that needs to be acknowledged is the sociological aspect... Informaticians can design all the systems they want, but unless those tools become useful and meaningful to the target users they are simply software and not informatics applications.

I would agree that Neuroinfomatics is a distinct discipline needing its own forum - along with bioinformatics, clinical informatics, public health informatics, and perhaps organized under an umbrella of Biomedical Informatics.

Answer 3

I always thought of the hierarchy as:

Informatics - the information infrastructure (software, data, databases, models, etc.) of everything

Under which is:

• Bioinformatics (the informatics of all the genetic and proteomic 'stuff')
• Clinical Informatics (all the electronic health record 'stuff')
• Neuroinformatics (the information infrastructure of neuroscience 'stuff')
• Probably lots of other *informatics for other topical fields and areas

It's a little unfortunate that the term 'bioinformatics' with the nicely broad 'bio' prefix ended up actually covering a fairly narrow (in the broad set of all informatics topics) set of topics, but when a group is the first to define a new term, you can do what you want, before the rest of the world catches up!

So, no, neuroinformatics is not a subset of Bioinformatics.

Answer 4

There are two common understandings of the term "neuroinformatics".
These two meanings are almost completely disjoint, but both are in common use.

1. Informatics applied to neuroscience:
   The application of information processing methods to data collected by neuroscientists.
      "How can we get the most out of our experimental data?"
   The challenges in this area are both organizational (promoting the availability and "databasing" of large datasets) and algorithmic (applying heavy-duty algorithms to these large datasets to extract useful information). The common theme is that computers and programs are powerful tools that should be used to help neuroscience.
   
     Some organizations using (primarily) this definition:

   • The International Neuroinformatics Coordinating Facility (including 16 national "nodes")
   • Frontiers in Neuroinformatics (journal)
   • This Q&A site, judging from the example questions suggested so far (by the first 7 contributors).
      

2. Informatics as carried out by neurons:
   The study of the information processing carried out by neurons in the brain.
      "How do neural signals relate to thinking?"
   The challenges in this area are to discover how populations of neurons (from single neurons up to large brain regions) represent and process information. The principles of neural information processing are largely unknown above the level of the single cell. Even foundational questions remain hotly debated, such as whether data is encoded by the precise timing of spikes or by the overall quantity of spikes. Researchers in this area attempt to discover the principles by which brains process information, and to reproduce brain-like behaviors using neural styles of computation.
   
     Some organizations using this definition:

   • The Institute of Neuroinformatics in Zurich
   • The Institute of Neuroinformatics in Bochum
   • The Neuroinformatics Group in Bielefeld

Of course many umbrella organizations take an inclusive approach and try to cover both 1 and 2.

Now, to answer your question:   "How is neuroinformatics NOT a subset of bioinformatics?"

Meaning 1 could be argued to technically fall under bioinformatics, although in practice there is no overlap in subject matter (see satra's answer).

Meaning 2 has nothing to do with bioinformatics.

In any event, as this site is still in the "definition phase", perhaps it would be worth explicitly addressing this issue of multiple meanings of the title of the site.