How to use hyphenation correctly in scientific manuscripts?

In other words: How does academic writing break grammar rules?

In this article you will learn:
Why does hyphenation matter?
The rules of correct hyphenation.
Are there exceptions?
How do you know if a word is common enough?
How should you decide in the difficult cases?
How can you check yourself what is the more common spelling?

When you are writing, or you could say carefully crafting, your manuscript, it goes without saying that you want to do your best and you certainly want to follow the rules of the English grammar.

Well, not so fast! Here are some surprising ways in which academic articles published in peer-reviewed journals break some of these rules.

Let’s take a closer look at hyphenation!

What is the big deal with hyphenation?

Why does using hyphens correctly matter? Whether or not you use a hyphen affects whether or not other scientists find your work. If others cannot find your work, your research won’t be read or cited [cue in career problems]. So let’s make sure you get those hyphens right.

Let’s take up-regulation and down-regulation as our examples.

I will let you know right here at the start that it is correct to write up-regulation and down-regulation and incorrect to write upregulation, up regulation, downregulation, or down regulation.

Here are the rules:

Prefixes that can stand alone as words (such as cross, half, up, and down) require hyphenation when used as adjectives; for example, cross-section, half-life, and all-inclusive.

Prefixes attached to a word or phrase starting with a capital letter (such as anti-HLA and non-Euclidean) should be hyphenated.

Prefixes that cannot stand alone as words (such as anti, bi, co, hyper, hypo, infra, inter, intra, micro, multi, peri, pre, pseudo, re, sub, supra, ultra) are not hyphenated when used as adjectives; for example, antimicrobial, biannual, cotransfection, hypervascular, and infrared. However, when these prefixes are used with two vowels or the same consonants abutting (such as hyper-reactive, pre-operative, anti-inflammatory, co-infection), then the term should be hyphenated.

Indeed, the correctly spelled up-regulation and down-regulation are also the official MeSH terms*.

However, a quick search on PubMed shows 203k search results for the incorrect upregulation and 291k for the correct up-regulation, while there are 166k search results for the incorrect downregulation and 163k for the correct down-regulation. You could say scientists get the hyphens right about half of the time.

If these numbers surprise you, you can also check Google Scholar. Here we get 1300k search results for the incorrect upregulation, 2370k for the correct up-regulation, 864k for the incorrect downregulation, and 2300k for the correct down-regulation. These results look better, although it is still astonishing how many times the rules get broken.

I do have to admit that I have been using the incorrect spelling of both words throughout my scientific career.

What is going on?

Let’s have one more look at PubMed search results but this time limiting our search to Nature as our golden standard [who doesn’t want to publish there?]. Here we get 149 search results for the incorrect upregulation, 112 for the correct up-regulation, 151 for the incorrect downregulation, and 177 for the correct down-regulation. These data are compatible with the more general search results from PubMed for these two terms.

Here are some data for you. The table contains a quick overview of how many results we get when searching for these words on PubMED and Google Scholar:

Word Nature
PubMED
Science,
PubMED
no journal,
PubMED
no journal,
Google Scholar
Rule
upregulation 149 0 203k 1300k incorrect
up-regulation 112 185 291k 2370k correct (can stand alone)
downregulation1518166k864kincorrect
down-regulation177188163k2300kcorrect (can stand alone)

First of all, it seems that Science is more careful with the editing their manuscripts than Nature. Perhaps we can conclude that scientists have been getting it wrong for so long that both spellings have become acceptable [some head scratching here]. In any case, my advice is to spell up-regulation and down-regulation correctly.

Does it mean there are exceptions to the rules?

Unfortunately, the answer is yes, there are exceptions. Words that are used commonly are not to be hyphenated, for example, outpatient and overexposed. So you have to be careful.

How do you know if a word is common enough?

Here is a quick summary of other commonly used words (with a number of search results in Nature on PubMed given in brackets):

WordNature
PubMED
Science,
PubMED
no journal,
PubMED
Rule
overexpression372208172kcorrect (exception)
over-expression9221kincorrect
immunohistochemistry9999391kcorrect (can’t stand alone)
immuno-histochemistry00717incorrect
chemoresistance1519905correct (can’t stand alone)
chemo-resistance20923incorrect
protooncogene4113kincorrect
proto-oncogene1068735187kcorrect (two vowels)
halflife110.3kincorrect
half-life11515591kcorrect (can stand alone)
antiinflammatory2118kincorrect
anti-inflammatory125116228kcorrect (two vowels)
antitumor112143127kcorrect (can’t stand alone)
anti-tumor5730kincorrect
antitumour7007kcorrect (can’t stand alone)
anti-tumour6806kincorrect
anticancer6677113kcorrect (can’t stand alone)
anti-cancer43228kincorrect
antiapoptotic3814kincorrect
anti-apoptotic25921kcorrect (two vowels)
proapoptotic73213kincorrect
pro-apoptotic301216kcorrect (two vowels)
prosurvival233kincorrect
pro-survival1744kcorrect (can stand alone)
preoperative27263kincorrect
pre-operative1025kcorrect (two vowels)
hyperreactive03989incorrect
hyper-reactive10294correct (same consonants)
costimulatory93310kcorrect (can’t stand alone)
co-stimulatory2524kincorrect
coinfection11916kincorrect
co-infection6110kcorrect (two vowels)

Based on our rule above, words like half-life, proto-oncogene, anti-inflammatory, immunohistochemisty, chemoresistance, and anticancer are used mostly correctly throughout the published literature. The word overexpression makes an exception due to common use.

The correct spelling of the word antitumor is also more common. Interestingly, the British spelling of the word antitumour/anti-tumour is more ambiguous with both terms almost equally in use. Likewise, prosurvival and pro-survival are used with an almost equal frequency.

The word anti-apoptotic is more commonly spelled correctly, not so the word pro-apoptotic.

Finally, we have the words preoperative or pre-operative, hyperreactive or hyper-reactive, coinfection or co-infection, and costimulatory or co-stimulatory. The journal Nature seems to prefer the hyphenated versions while Science prefers not to hyphenate. Generally, the incorrect spelling (preoperative, hyperreactive, and coinfection) seems to be more common, so perhaps we are working towards other exceptions to the rule.

What should you do? How should you write these words?

My advice is to follow the rules. You could use the incorrecly spelled word if it is clearly used more frequently in the published literature.

The reason for my advice is to do what the most scientists do so they can find your articles and cite your research which is good for your career.

How can I check myself what is the more common spelling?

Fortunately, checking the more common spelling is very easy and you only need to do it once. You can also check specifically the preferences of your target journal, because we have seen that, for example, Nature and Science have different spelling preferences.

Here is how you do that:

  1. Go to https://www.ncbi.nlm.nih.gov/pubmed.
  2. Search for: “overexpression” AND “J Neurosci”[Journal]
  3. Search for: “over-expression” AND “J Neurosci”[Journal]

Remember to put quotation marks “” around your journal if it has more than one word.

There you go. Enjoy your writing and good luck!

* MeSH (Medical Subject Headings) is the National Library of Medicine’s controlled vocabulary thesaurus, used for indexing articles for the MEDLINE┬«/PubMED┬« database. Each article citation is associated with a set of MeSH terms that describe the content of the citation.”