By: David Shifrin, PhD Science Writer, Filament Life Science Communications
We all know the feeling: A new project starts, so we set up a folder system and naming convention and get to work. It works pretty well for a while; most of the files fit into the folder hierarchy on our computer and the names generally make sense. The occasional outlier can be shoehorned into the existing scheme. But then, after some period of time, we go to look for a file and can’t find it. Or, we get a new file that just makes no sense with the naming convention. Or, the folder names and organization don’t work anymore because the project has become too complicated.
What do we do? Take a day and reorganize everything, creating a new, more complex convention on our desktop? Or, avoid rebuilding and hope the computer’s search function will make up for our disorganization?
This type of situation isn’t limited to our individual work. Often, including science and medicine, it can happen across an entire field. In those instances, not only does the re-organization require a significant effort just to come up with a workable system, but also the many individuals working within that field have to come on board or risk even more confusion.
Enough with the hypotheticals, let’s look at an existing and current example in the world of genetic diseases. Charcot-Marie-Tooth disease is a peripheral neuropathy that affects ~1/2500 people in the US, according to the NIH. Those numbers make it one of the more prevalent neurological disorders. The issue, though, is that CMT is really a large set of distinct conditions that all have similar phenotypes but different genetic etiologies. Well over 30 genes have been directly implicated in CMT, and roughly twice that have been associated with the disorder(s) and related conditions. The modes of inheritance vary, virtually every type of mutation is represented (duplication, mutation, etc.), and the way in which peripheral nerves are affected can be different (e.g., axonal defects vs demyelination). As a result, classifying and subdividing CMT as new syndromes are discovered is a complex task.
Historically, the simple (and common) method of classification has been used: tack a number and letter to the end and call it good. Currently, the field recognizes three main types of CMT, CMT1-4. Each type contains between 5 and 7 subtypes. While “CMT1F” is convenient, it’s arguably no longer sufficient to truly describe the syndrome it represents.
To solve this problem, a group of scientists from several neurology departments across France and Algeria has suggested a new nomenclature for CMT. They begin with an impressively comprehensive review of CMT, its history and many genotypes. Informative in itself, this review also highlights the remarkable level of heterogeneity within the umbrella categorie of CMT neuropathies.
From there, the authors get into the practical implications regarding nomenclature. They point out that different classification systems almost match up, but “almost” isn’t good enough. For example, CMT1, 2 and 3 are the same as Hereditary Motor and Sensory Neuropathy (HMSN) I, II and III. However, CMT4 “[…] does not correspond to HMSN IV,” but instead matches a separate condition. Simply put, the correlation is not linear.
The review covers a number of other problems. What it all comes down to, though, is the statement that “the discovery of many genes susceptible to be responsible […] made the classification of CMT more and more complicated.” Additionally, there is still some flexibility in classification within certain CMT subtypes.
With all that in mind, the authors suggest a new naming convention that is more complicated on the surface but clears up much of the ambiguity surrounding Charcot-Marie-Tooth. The first step is to abbreviate the mode of inheritance – AD for autosomal dominant, AR for autosomal recessive, XL for X-linked, etc. Various forms of CMT should then be reorganized by key phenotype, such as demyelination. The examples given are CMT1 (autosomal dominant demyelinating) and CMT4 (autosomal recessive demyelinating), which would be reclassified as AD-CMT1 and AR-CMT1, respectively. To clarify even further, the authors consider the possibility of removing numbers altogether and replacing them with abbreviations of the phenotype. This would convert CMT1 into CMTde (and “ax” for axonal). The last step would be to provide maximum granularity by replacing the generic alphabetic labels with the actual gene involved. Therefore, the authors propose, “CMT1B and CMT2I/J could be named AD-CMTde-MPZ and AD-CMTax-MPZ, respectively.”
Again, this system becomes significantly more complex in terms of writing out the names of various syndromes. But, this minor inconvenience would be more than balanced by the clarity provided by spelling out the specifics of each subtype. There would be less information to remember, which means that clinicians could review data faster and with less confusion. Additionally, the group says, “these changes would probably be more readable for non-experts […]” which may be a worthy goal for making genetics more accessible.
Certainly, many fields have put together naming convention updates. However, it seems likely that this will become more necessary more often. Even if we are not finding a significant number of new diseases in phenotypic terms, the rapid accumulation of genetic data is revealing the complexity and heterogeneity that underlies many known syndromes. More granular information will require reclassification to help with both descriptive accuracy and therapeutic processes.