We welcome the comments of Dr Bandelt et al on studies analyzing mtDNA in relation to the genetic bases of schizophrenia, especially those focusing on our study.

With respect to the putative technical errors that they mentioned, which could have influenced our RFLP analysis, we need to highlight that in our report, we had stated that all samples have the same concentration of DNA (10 ng/μl), spectrophotometrically measured, and that a positive control (i.e. the patient carrying the variant) was used in each analysis. Therefore, these potential problems impacting on RFLP analysis have been controlled-for in our study. No doubt, the possibility that the single occurrence of the T12096A could have spread to other samples via cross-contamination needs to be considered, even though the genetic material we analyzed was derived from 10 ml of peripheral blood samples collected and processed at different times of the study.

Patients were assigned to haplogroups H, K and W based on the presence of the variants G1719A, G4580A, C7028T, G9055A, A12308G, G13368A, G13708A and G16391A, as Huerta et al1 had recently described in 250 Spanish patients diagnosed as having Parkinson's disease, and in 230 healthy individuals. Taking into account these eight variants, we could assign each sample to one unequivocal haplogroup. Unfortunately, we did not perform any phylogeny-based approaches in the analyses.

We agree that the DNA sequences of the schizophrenia patients were initially aligned to the Swedish sequence X93334 gi 1262342 and, then, only the discrepancies between the two sequences being compared to the rCRS2 in the Human Mitochondrial Genome Database (www.mitomap.org). The alignment of the rCRS and the X93334 reveals that X93334 has 36 variant nucleotides from the rCRS, and 17 of them are present in the schizophrenia patients. This explains why some variants have been overlooked in our report. Table 1 (below) shows these 36 nucleotide variants in each of the six schizophrenia patients.

Table 1

Additionally, in our article, we did not mention that the new mitochondrial DNA variants are related to the disease outcome, as Dr Bandelt et al indicate. Neither, would we claim that the variants we found in ‘only’ six mother–offspring pairs of schizophrenia patients would raise major diagnostic and clinical expectations. Indeed, we had indicated that further studies in larger samples are needed (Discussion section) and that our results needed to be validated and replicated in many more patient samples (Discussion section).

Unfortunately, some of the recommendations from Dr Bandelt et al3, 4 have been drawn from very recent articles and books, which were unavailable to us at the time of our study. Undoubtedly, in future studies, we will incorporate the data-mining strategies suggested.