On 2015 Apr 21, Bernard J Crespi commented:

We are happy to clarify some points regarding our article, and reply to the issues raised in the comment from Dr. Bishop.

First, we would consider the title of our article to be accurate, as it simply refers to mediation of schizotypy and handedness by the LRRTM1 gene, and does not, as the comment indicates, make any statements about SNPs in relation to handedness. We consider epigenetic phenomena, such as the methylation that we measured for LRRTM1, to be important mediators of a gene's functional effects. For example, Brucato et al. (2014) have recently linked methylation of LRRTM1 with risk of schizophrenia.

Second, we used a measure of strength of handedness, which is compatible with Francks et al. (2007) only in being a continuous measure. We apologize in that we could have worded the relevant sentence more precisely. We did not use a specific measure that, like that of Francks et al. (2007), may confound handedness direction with handedness strength, because these two variables, direction and strength, appear to exhibit independent genetic underpinnings (e. g., Ocklenberg et al. 2014). Moreover, schizophrenia and schizotypy show associations predominantly with handedness strength, not direction, from previous work (e. g., Chapman et al. 2011).

Third, we found a significant association at P = 0.026 (r = -0.375, product-moment correlation) for an association of PC1 (a composite methylation score) with handedness for our full sample, based on our predictions. We also provided the results for males and females separately, because of extensive evidence of gender differences for cognitive and psychiatric phenotypes such as those analyzed here. Considered separately, only the results for females were statistically significant (at r = -0.469, P = 0.032). Higher statistical significance would always be more compelling, of course.

We welcome insightful comments and constructive criticism, and we hope that our results regarding LRRTM1 will motivate further research into the effects of genetic and epigenetic variation in this fascinating gene.

References

Brucato N, DeLisi LE, Fisher SE, Francks C. Hypomethylation of the paternally inherited LRRTM1 promoter linked to schizophrenia. Am J Med Genet B Neuropsychiatr Genet. 2014 Oct;165B(7):555-63.

Chapman HL, Grimshaw GM, Nicholls ME. Going beyond students: an association between mixed-hand preference and schizotypy subscales in a general population. Psychiatry Res. 2011 May 15;187(1-2):89-93.

Francks C, Maegawa S, Laurén J, … Monaco AP. LRRTM1 on chromosome 2p12 is a maternally suppressed gene that is associated paternally with handedness and schizophrenia. Mol Psychiatry. 2007 Dec;12(12):1129-39, 1057.

Ocklenburg S, Beste C, Arning L. Handedness genetics: considering the phenotype. Front Psychol. 2014 Nov 11;5:1300.

On 2015 Apr 19, Dorothy V M Bishop commented:

The genetics of handedness is a topic where there seems to be a considerable mismatch between conventional wisdom, where handedness is regarded as a highly heritable trait, and research, which has struggled to find evidence of significant heritability. I am aware that a meta-analysis of twin studies by Medland and colleagues reported heritability of around .25, but a subsequent study by Armour et al (2014) with N = 3940 failed to find any locus that reached genome-wide significance. I also appreciate that this does not rule out genetic influences with small effect sizes, especially when the picture is made more complex by potential imprinting.

I think, though, that the title of this paper does rather overstate the case in claiming that LRRTM1 mediates schizotypy and handedness. (My focus is solely on the claims regarding handedness).

I appreciate that researchers in this area have some major problems to grapple with. One is that there is no agreement about how to define the phenotype of handedness. Most studies rely on handedness questionnaires, but these have only an imperfect relationship with measures of relative hand skill. In the absence of an adequate underlying model of etiology of handedness, almost any kind of measure can be justified: a binary distinction between left and right, a distinction between left, mixed and right, a continuous, quantitative scale, or a distinction between consistent (left or right) vs inconsistent (mixed) handers. As I argued many years ago, such flexibility is dangerous, because it almost always possible to find some handedness measure that will show an apparently meaningful relationship with a criterion measure (Bishop, 1990).

In this study, the handedness measure was strength of handedness, assessed by "taking the absolute value of the handedness score such that values near zero represent mixed handedness and values near 64 represent strong handedness (right or left)." It is stated that this was done to be consistent with the previous study on LRRTM1 by Francks et al (2007). However, my impression was that Francks et al used a continuous quantitative phenotype that represented relative hand skill, with R>L at one end and L>R at the other. Thus the raw laterality quotient (-64 to +64) from the Waterloo Handedness Questionnaire would seem conceptually closer to the measure used by Francks et al than the absolute measure. It's possible I have misunderstood this (Francks et al is an exceedingly complex paper), but I'd be grateful for clarification.

I find the title misleading because in the current paper the LRRTM1 SNPs were not associated with the absolute handedness measure used with this sample. (This cannot be regarded as a failure to replicate Francks et al, since parent-of-origin effect were not analyzed and the measurement of handedness appears to have been different). The reference in the title to the gene mediating handedness presumably refers to the significant correlation between methylation in CpG sites and absolute handedness. However, this result does not survive correction for multiple comparisons. I therefore would suggest that the evidence for an association between LRRTM1 and handedness in this study is not compelling.

References

Armour, J. A. L., Davison, A., & McManus, I. C. (2014). Genome-wide association study of handedness excludes simple genetic models. Heredity, 112(3), 221-225. doi: 10.1038/hdy.2013.93

Bishop, D. V. M. (1990). How to increase your chances of obtaining a significant association between handedness and disorder. Journal of Clinical and Experimental Ñeuropsychology, 12, 786-790.

Francks, C., Maegawa, S., Lauren, J., Abrahams, B. S., Velayos-Baeza, A., Medland, S. E., . . . Monaco, A. P. (2007). LRRTM1 on chromosome 2p12 is a maternally suppressed gene that is associated paternally with handedness and schizophrenia. Molecular Psychiatry, 12(12), 1129-1139.

Medland, S. E., Duffy, D. L., Wright, M. J., Geffen, G. M., Hay, D. A., Levy, F., . . . Boomsma, D. I. (2009). Genetic influences on handedness: Data from 25,732 Australian and Dutch twin families. Neuropsychologia, 47(2), 330-337. doi: 10.1016/j.neuropsychologia.2008.09.005

On 2015 Apr 19, Dorothy V M Bishop commented:

The genetics of handedness is a topic where there seems to be a considerable mismatch between conventional wisdom, where handedness is regarded as a highly heritable trait, and research, which has struggled to find evidence of significant heritability. I am aware that a meta-analysis of twin studies by Medland and colleagues reported heritability of around .25, but a subsequent study by Armour et al (2014) with N = 3940 failed to find any locus that reached genome-wide significance. I also appreciate that this does not rule out genetic influences with small effect sizes, especially when the picture is made more complex by potential imprinting.

I think, though, that the title of this paper does rather overstate the case in claiming that LRRTM1 mediates schizotypy and handedness. (My focus is solely on the claims regarding handedness).

I appreciate that researchers in this area have some major problems to grapple with. One is that there is no agreement about how to define the phenotype of handedness. Most studies rely on handedness questionnaires, but these have only an imperfect relationship with measures of relative hand skill. In the absence of an adequate underlying model of etiology of handedness, almost any kind of measure can be justified: a binary distinction between left and right, a distinction between left, mixed and right, a continuous, quantitative scale, or a distinction between consistent (left or right) vs inconsistent (mixed) handers. As I argued many years ago, such flexibility is dangerous, because it almost always possible to find some handedness measure that will show an apparently meaningful relationship with a criterion measure (Bishop, 1990).

In this study, the handedness measure was strength of handedness, assessed by "taking the absolute value of the handedness score such that values near zero represent mixed handedness and values near 64 represent strong handedness (right or left)." It is stated that this was done to be consistent with the previous study on LRRTM1 by Francks et al (2007). However, my impression was that Francks et al used a continuous quantitative phenotype that represented relative hand skill, with R>L at one end and L>R at the other. Thus the raw laterality quotient (-64 to +64) from the Waterloo Handedness Questionnaire would seem conceptually closer to the measure used by Francks et al than the absolute measure. It's possible I have misunderstood this (Francks et al is an exceedingly complex paper), but I'd be grateful for clarification.

I find the title misleading because in the current paper the LRRTM1 SNPs were not associated with the absolute handedness measure used with this sample. (This cannot be regarded as a failure to replicate Francks et al, since parent-of-origin effect were not analyzed and the measurement of handedness appears to have been different). The reference in the title to the gene mediating handedness presumably refers to the significant correlation between methylation in CpG sites and absolute handedness. However, this result does not survive correction for multiple comparisons. I therefore would suggest that the evidence for an association between LRRTM1 and handedness in this study is not compelling.

References

Armour, J. A. L., Davison, A., & McManus, I. C. (2014). Genome-wide association study of handedness excludes simple genetic models. Heredity, 112(3), 221-225. doi: 10.1038/hdy.2013.93

Bishop, D. V. M. (1990). How to increase your chances of obtaining a significant association between handedness and disorder. Journal of Clinical and Experimental Ñeuropsychology, 12, 786-790.

Francks, C., Maegawa, S., Lauren, J., Abrahams, B. S., Velayos-Baeza, A., Medland, S. E., . . . Monaco, A. P. (2007). LRRTM1 on chromosome 2p12 is a maternally suppressed gene that is associated paternally with handedness and schizophrenia. Molecular Psychiatry, 12(12), 1129-1139.

Medland, S. E., Duffy, D. L., Wright, M. J., Geffen, G. M., Hay, D. A., Levy, F., . . . Boomsma, D. I. (2009). Genetic influences on handedness: Data from 25,732 Australian and Dutch twin families. Neuropsychologia, 47(2), 330-337. doi: 10.1016/j.neuropsychologia.2008.09.005

On 2015 Apr 21, Bernard J Crespi commented:

We are happy to clarify some points regarding our article, and reply to the issues raised in the comment from Dr. Bishop.

First, we would consider the title of our article to be accurate, as it simply refers to mediation of schizotypy and handedness by the LRRTM1 gene, and does not, as the comment indicates, make any statements about SNPs in relation to handedness. We consider epigenetic phenomena, such as the methylation that we measured for LRRTM1, to be important mediators of a gene's functional effects. For example, Brucato et al. (2014) have recently linked methylation of LRRTM1 with risk of schizophrenia.

Second, we used a measure of strength of handedness, which is compatible with Francks et al. (2007) only in being a continuous measure. We apologize in that we could have worded the relevant sentence more precisely. We did not use a specific measure that, like that of Francks et al. (2007), may confound handedness direction with handedness strength, because these two variables, direction and strength, appear to exhibit independent genetic underpinnings (e. g., Ocklenberg et al. 2014). Moreover, schizophrenia and schizotypy show associations predominantly with handedness strength, not direction, from previous work (e. g., Chapman et al. 2011).

Third, we found a significant association at P = 0.026 (r = -0.375, product-moment correlation) for an association of PC1 (a composite methylation score) with handedness for our full sample, based on our predictions. We also provided the results for males and females separately, because of extensive evidence of gender differences for cognitive and psychiatric phenotypes such as those analyzed here. Considered separately, only the results for females were statistically significant (at r = -0.469, P = 0.032). Higher statistical significance would always be more compelling, of course.

We welcome insightful comments and constructive criticism, and we hope that our results regarding LRRTM1 will motivate further research into the effects of genetic and epigenetic variation in this fascinating gene.

References

Brucato N, DeLisi LE, Fisher SE, Francks C. Hypomethylation of the paternally inherited LRRTM1 promoter linked to schizophrenia. Am J Med Genet B Neuropsychiatr Genet. 2014 Oct;165B(7):555-63.

Chapman HL, Grimshaw GM, Nicholls ME. Going beyond students: an association between mixed-hand preference and schizotypy subscales in a general population. Psychiatry Res. 2011 May 15;187(1-2):89-93.

Francks C, Maegawa S, Laurén J, … Monaco AP. LRRTM1 on chromosome 2p12 is a maternally suppressed gene that is associated paternally with handedness and schizophrenia. Mol Psychiatry. 2007 Dec;12(12):1129-39, 1057.

Ocklenburg S, Beste C, Arning L. Handedness genetics: considering the phenotype. Front Psychol. 2014 Nov 11;5:1300.