About this blog

Wednesday, May 16, 2012

Is MTMR2 Causing CMT4?


A lot has happened over the weekend. I've received some very encouraging e-mails as more people find out about this project. I want to thank you all for your interest. I wasn't sure what to expect, but my hope is still to make this as accessible and easy as possible to remove some of the "aura of mystery" surrounding genetic analysis. We live in an era when we can go from receiving a digital file through the internet to literally searching for the genetic basis of a mendelian disease in a loved one. While manually inspecting variants should not replace expert analysis, it can only help. In fact, Stacy initially went to the hospital to determine the genetic cause of her disease. She went to the hospital and they told us that they would check only known genetic causes for CMT4 for ~$3,000 and that insurance wouldn't cover the cost. This is about the same time that I heard 23andMe was about to launch a pilot for sequencing 80x exomes for $1000. Stacy decided to wait for this and then have me personally analyze her exome. This blog is the result of that decision. We are on the brink of an era when full genome sequences for all of us will be common place. It is my hope that in the future, people will not have to wonder what genetic mutation is causing their mendelian disease. Even if the result is not currently actionable, they will at the very least know what technological advances need to be made before a symptom decreasing therapeutic intervention is developed.

I'm showing how all of you can join in and analyze Stacy's exome for known causes of CMT4. In addition, I will show how to search for novel mutations that are likely causing the disease if it isn't explained by a known cause. There are many who believe what we do is shrouded in mystery, but there is a clear list of things that should be attempted by anyone with a rudimentary understanding of genetics, is handy with a computer, and who (as I do) has an unlimited amount of motivation. I am doing this because Stacy was about to pay ~$3,000 to analyze 7 genes for known mutations. Instead, she decided to pay $1,000 and have me analyze thousands of genes as I will look for further mutations unrelated to CMT4.

With no further delay, we will now be moving on to the 4th gene to test for CMT4 causing mutations. We have now excluded known causal mutations FGD4FIG4, and GDAP1. We have not found the causal mutation and we have a few genes left to check before we conclude that there was a misdiagnosis or that there is a novel mutation. We will now check MTMR2 for CMT4 causing mutations in Stacy's exome...


This previous post explains how to visualize a gene. Apply this to Stacy's copy of MTMR2 if you don't know how to browse through a bam file to analyze individual mutations.

Below is a list of CMT4 causing mutations from the HMDG and the IGN mutation website. We will examine each CMT4 causing mutation in Stacy's copy of MTMR2. I will take a screenshot of anything of interest and discuss potential ramifications.

We have excluded 3 genes and have moved on to the 4th.
  1. FGD4  -- 5 mutations cause CMT4
  2. FIG4 -- 20 mutations cause CMT4
  3. GDAP1 -- 34 mutations cause CMT4
  4. MTMR2 -- 10 mutations cause CMT4
  5. PRX -- 2 mutations cause CMT4
  6. SBF2 -- 5 mutations cause CMT4
  7. SH3TC2 - 19 mutations cause CMT4

Before examining MTMR2, make sure to right-click the box containing the text "RefSeq Genes" on the lower left of the IGV viewer. Then select "Expanded" as opposed to "Collapsed". You should now see 4 different MTMR2 variants at the bottom of the screen. These 4 different variants of the same gene exist due to alternative splicing. Since the exons are different, the codon numbers also differ across different versions of the gene. For our purposes, you can safely ignore the top and bottom two copies. The following analysis applies to the second copy of MTMR2 from the top.

There are two things to keep in mind when analyzing this gene.
  • This gene is expressed from the "right to left", so we're looking for each codon in reverse order (i.e. CCA instead of ACC).
  • **A codon may appear not to match the expected amino acid. If this occurs, please click "Sequence" text on the lower left side of the IGV viewer and click "Flip Strand".
    • After flipping the stand, each codon is read in reverse (3rd position is now 1st position, etc.). In addition, you must reverse complement the base reported in the Stacy's reads. For example, a T becomes an A, a C becomes a G, etc.


Mutations causing CMT4 in the MTMR2 gene.
  1. Gly103Glu Matches Reference Genome
  2. ?107? DELETION Matches Reference Genome
  3. ?103? INSERTION Matches Reference Genome
  4. Thr108fs Matches Reference Genome
  5. Thr228fs Matches Reference Genome
  6. Glu276X Matches Reference Genome
  7. Ile281fs Matches Reference Genome
  8. Arg283Trp Matches Reference Genome
  9. Gln426X Matches Reference Genome
  10. Arg477X (Did they mean 476?, 477 is Ser and 476 is Arg). Either way, they both  Match the Reference Genome
  11. Gln482X Matches Reference Genome
  12. Phe494Glu Matches Reference Genome
  13. Glu531del Matches Reference Genome
  14. ?578? INSERTION/DELETION Matches Reference Genome
  15. Tyr579fs Matches Reference Genome
  16. Trp583X INSERTION/DELETION Matches Reference Genome
Mutation 10 likely contains an error. Codon 477 is Arg and codon 476 is Ser. However, the mutation that is known to cause CMT4 at this position is reported to be Arg477X. My guess is that there was a mistake when  476 and 477 was typed instead. To be safe, I checked position 477 in alternatively spliced variants and everything matches the reference genome. Stacy does not contain any known CMT4 causing mutations in MTMR2.


We have now eliminated all known CMT4 causing mutations in 4 genes. I couldn't help myself, and I quickly reviewed all genes for regions that fail to match the reference genome. I have identified what I consider to be a very likely novel mutation that may be causing CMT4 in Stacy. However, I am determined to be methodical and finish checking all known CMT4 mutations as well as all known CMT causing mutations (CMT1, CMT2, etc.) in case there was a misdiagnosis. Only then will I seriously consider novel mutations.

In the next post, we will examine PRX for CMT4 causing mutations...

No comments:

Post a Comment