It appears that the authors’ intent is to show inherent karyotypic instability in pluripotent cell lines, but there is no evidence in this report that they carefully controlled for conditions that could result in erroneous gain and loss of chromosomes in metaphase preparations. There is also no evidence that the authors used established resources for interpretation of results of FISH analysis or clonality. Though they implicate the blastocyst (it is important to distinguish a blastocyst from an embryonic stem cell line) as a source of potential cytogenomic variability, the iPS line that they examined also veered from the modal number of 46. There is established bias toward recurrent chromosome gain in human pluripotent stem cells (see for example: Baker et al., Nat. Biotechnol. 25, 207-215, 2007 and Taapken et al., Nat. Biotechnol 29, 313-315, 2011); in this study, there appears to be a bias toward loss, consistent with preparation artifact and analytical error.

The website of the American College of Medical Genetics (http://www.acmg.net/) gives open access to the well-vetted Standards and Guidelines for Clinical Genetics Laboratories E. Clinical Cytogenetics. This document is based on years of experience with many types of specimens; it results in best practices for clinical diagnostics to serve patients. Stem cell researchers, with or without potential clinical applications, deserve the same careful attention.

Points to consider from these guidelines:
1. In section E3.1.1, terminology is described:
“Clone is defined as a cell population derived from a single progenitor cell. Clonal origin is inferred by the presence of at least two cells containing the same extra chromosome(s) or structural chromosome abnormality or by the presence of at least three cells that have lost the same chromosome.
For the purpose of constitutional studies, the use of the terms cell line and clone are interchangeable.
Mosaicism is the presence of two or more cytogenetically distinguishable cell lines.
Pseudomosaicism refers to the presence of an abnormal cell(s) in cultured cells that arise from an in vitro culture artifact and do/does not represent the true karyotype.”

a. Did the authors attempt to establish clonality, i.e., true mosaicism, by counting chromosomes in additional cells? In Table 2, the random loss or gain per culture, though mathematically correct at 2.5%, is actually one of forty cells examined.
b. The authors’ discussion related to “constitutive aneuploidy” is confusing. Do they mean to write “constitutional”? If so, this would imply clonality. Do they suggest that recurrent aneuploidy in pluripotent stem cells is not acquired?

2. Interphase FISH analysis has inherent hybridization variability and also requires understanding of the visualization of three dimensional nuclei in two planes. Correct interpretation of probe signals requires careful attention to existing guidelines. It does not appear that the authors determined analytical sensitivity and/or reportable ranges for the FISH probes that they used; without validation, FISH assay comparisons are invalid. For description of FISH technical standards and guidelines, please see Mascarello et al., Section E9 of the ACMG technical standards and guidelines: FISH. Genetics in Medicine 13, 667-676, 2011; and Dewald et al., A multicenter investigation with interphase fluorescence in situ hybridization using X- and Y-chromosome probes. AJMG 76:318-316, 1998. Did the authors use any of these or other processes to assess probe performance?

a. The authors state that their “error rate” is < 3/10,000 in lymphocytes. Within Figure 4, an “error rate” is listed as 1/10,000, but it isn’t clear what cell type. Is this value attributed to the probe manufacturer or was this rate derived by the authors? How? By error rate, do they imply false negatives and positives? Does this assess the performance of the probes in their cell cultures?
b. Were the probe signals for both chromosomes scored together? There is no table to describe or document the FISH analysis and any validation assays that may have been done in this study.

Other questions that arise upon review of this publication:
1. The use of DAPI staining and SKY is usually reserved for situations where the chromosomes are too small to be clearly identified by G-banding, or to determine complex structural rearrangements. In this report (e.g., Figure 3A and B), the chromosomes are quite condensed. We rarely see this type of morphology in metaphase preparations from pluripotent stem cell cultures. In this study, there appears to be a bias toward loss of chromosomes from F and G groups (the smallest chromosomes of the human karyotype)—of the nineteen cells with missing chromosomes, ten have lost chromosomes 19, 20, 21, and 22. These chromosomes may be the most likely to float away if spreading conditions are not carefully controlled. Stating that metaphases do not look overblown or have trailing chromosomes is not sufficient. How many fields of view away from the scored metaphases did the authors search? Uneven illumination, as evidenced in Figure 1B, can also result in errors in chromosome counting. Is Figure 1B representative of the lighting conditions that the authors used for chromosome counting?
2. Lymphocytes, obtained from short term culture, were used as a control. Would the authors have come to the same conclusions had they compared their results to a non-pluripotent, established cell line in long term culture?
3. The authors do not detail the condition of metaphase chromosome preparations: were the conditions (temperature, humidity, air flow) of spreading, for cell lines and for lymphocytes, consistent and well-documented?
4. Although multiple laboratories cultured the cell lines using different methodology, the authors don’t reveal whether the metaphase preparations were made in more than one lab. It appears that the only side by side comparison for reproducibility of results was that of the H9 cell line that was prepared in our lab; our preparation did not result in random chromosome loss or gain.


Insight into the basis for karyotypic aberrations in pluripotent stem cells is a critical need that requires accurate use of the tools of cytogenetic analysis. There are many resources available to assure accurate interpretation of cytogenetic assays in order to differentiate art from artifact.