Significance of Variants

Now that we are able to get the genetic sequence from a small cell sample such as a cheek swab, a blood sample or a piece of tumour tissue, we can compare its sequence to the reference human genome, a standardised sequence for a ‘typical’ human.

Using bioinformatics, fragments of sequence are aligned to the reference and changes are identified. There are several types of changes we might find in a genome sequence, and these changes may or may not be clinically important. Classification of variants is important in both clinical and commercial testing; recall the discussions of genetic risk in week 1.

Variants can be any of the following:

• simple point changes known as single nucleotide variants (SNVs)
• insertions and deletions (of variable size)
• translocations (a chunk of genetic material moved to a new location)
• copy number variants (duplications, triplications etc)
• structural variants (variation in modifications which affect the way the DNA is folded and hence accessed)

In addition, the changes can lie in exons (coding regions) or intronic areas. A typical individual European genome contains 3 million variants from the reference human genome so we have to do some serious bioinformatic filtering to narrow this down to clinically relevant information.

There are three main categories of variants; the vast majority of the variants are common and benign. A smaller number are known pathogenic variants and then there is a whole tranche of variants of unknown significance.

It is also important to realise that during the multiple steps of sequencing and processing of the data, errors can be generated which quality control steps try to eliminate. It is a bit like the error reading on a blood test – the answer is not absolute and a potential variant has to be evaluated and triangulated as many different ways as possible. This might be carried out for example by re-sequencing the region using Sanger sequencing, by checking if the variant is seen in other species, by predicting function, by testing function in vitro and by looking at patterns of inheritance.

In a report obtained via a clinical genetics service, a targeted question will be being asked such as ‘is there a mutation present in a gene known to be associated with a particular disease?’ The answer will be given along with an assessment of the accuracy and sensitivity of the testing process. New variants will have been assessed using bioinformatic in silico (computer based) and in vitro testing to determine the likelihood of pathogenicity and, in general, variants of unknown significance are not reported.