Python Programming for Biology: Bioinformatics and Beyond

Conservation and variation

In earlier chapters we have analysed sequences to detect their similarities and thus to form

alignments.  The  purpose  of  alignment  is  not  only  to  group  sequences,  but  also  to  say  in

what  precise  ways  the  sequences  differ,  or  are  preserved.  If  we  look  down  a  column  of

letters  along  the  various  positions  of  a  multiple  alignment  some  locations  will  use  more

residue types than others, and if we are considering protein sequences we can see places

where the chemical character of the amino acids may remain the same despite the precise

residues being different. Accordingly, we can measure how conserved or variable a given

position is. Combining many positions we can say how variable a whole region or whole

gene  is.  We  can  analyse  sequence  variability  and  find  changes  of  biological  or  medical

importance, and also learn something about the evolution and origins of the sequences.

As discussed, when organisms reproduce, sequence changes naturally occur. However,

not  all  changes  in  the  DNA  are  of  consequence  and  many  have  no  effect  at  all  on  an

organism. This is because not all DNA has an immediate biological role, and even within

the regions that do there can often be several sequences that perform a job equally well.

Generally,  changes  in  DNA  which  are  not  important  for  biological  function  occur  more

frequently; there is no reason for them not to be passed on.

In the genomes of many organisms there is a high proportion of non-functional, often

repetitive  junk  DNA  between  genes.  This  is  not  to  say  that  all  DNA  between  genes  is

useless, given that such intergenic regions must contain control elements to regulate gene

expression, promoters and enhancers, and also structural DNA to maintain chromosomes,

like  telomeres  to  protect  chromosome  ends  and  centromeres  to  allow  replication.  Some

regions of non-functional DNA tend to show the highest rate of change during evolution

and  thus  the  largest  variation  between  individuals  of  a  species.  Human  DNA

fingerprinting,  for  example,  which  may  be  used  to  identify  criminals  or  detect  family

members,  works  by  looking  at  hypervariable  regions  that  are  different  in  almost  every

person.  Such  fingerprinting  would  not  work  nearly  so  well  if  gene-coding  regions  were

used;  there  would  be  far  fewer  differences  and  finding  two  individuals  with  the  same

sequence  (i.e.  not  the  real  criminal)  would  be  much  more  likely,  and  in  some  cases

positively expected.

The  task  of  some  analyses,  rather  than  to  just  detect  variations,  is  to  measure  the

relative rate of change of variation. If we can find sites where the rate of change of the

sequence is above or below the normal expected value, then this tells us something about

the  process  of  evolution  at  a  fine  scale.  A  common  rate  measure  for  variations  in  the

coding  regions  of  genes  that  go  on  to  make  protein  is  to  look  at  the  number  of  DNA

substitutions that do change the amino acid sequence, compared to those that do not; the

synonymous, silent substitutions. Remember that the number of three-base codons (64) is

larger than the number of amino acids, and there are usually different ways of coding for

the same amino acid.

---

In  regions  that  have  more  silent  changes  than  active  ones  the  acceptance  of  the

sequences during evolution indicates a purifying selection; this sequence is important and

there is a reason why the protein sequence is preserved. Where there are proportionately

more  active  changes  than  silent  ones,  compared  to  the  average,  then  this  can  indicate  a

region where there is positive selection. Such regions indicate that the rate of evolution at

these  sites  is  greater  than  normal  and  that  continuous  change  and  adaptation  is

advantageous. Regions of positive selection in the human genome include genes involved

in the immune system, which are ever changing to cope with the continuous appearance of

new harmful bacteria, viruses and parasites.