2. Theoretical ways in which false‐positive and false‐negative results could be obtained from the various techniques.

Technique	Potential ways false‐positive results could be obtained	Potential ways false‐negative results could be obtained
FISH	Focal deletions at regions that the target probes hybridise could lead to false‐positive results as these cannot be distinguished from whole arm deletions (as only 1 probe per chromosome arm is normally used).	False‐negative results could be obtained if there has been a loss of heterozygosity without copy number reduction.
	Depending on the way that deletions are diagnosed (i.e. the cut‐off used and whether it depends on the ratio of test probes to control probes), aberrations that lead to disproportionate gain in control probe loci (i.e. 1q and 19p) could lead to false‐positive results.	False‐negative results could be obtained if non‐neoplastic nuclei are assessed.
	The way that the tumour tissue is sectioned to prepare it for FISH could lead to 'truncation artefact'. Nuclei may be transected, which may lead to them containing incomplete genetic material. False‐positive results may be obtained from normal tissue in the presence of excessive truncation artefact.	Excessive truncation artefact in neoplastic tissue could lead to false‐negative results.
CISH	As for FISH.	As for FISH.
PCR‐based LOH assays	PCR cannot distinguish between relative and absolute deletions, so people with relative deletions will be given false‐positive results.	If tumour samples are heavily contaminated with normal tissue, PCR products for both alleles will be obtained in a ratio that would give a false‐negative result.
	Depending on primer spacing and the number of informative loci, the technique may detect focal rather than whole arm deletions.
	Imbalanced polysomy, e.g. gain of 1 copy of chromosome 1 and 19, may result in allelic imbalance and be interpreted as loss of heterozygosity.
RFLP analysis	Cannot distinguish between relative and absolute deletions, so people with relative deletions will be given false‐positive results.	If tumour samples are heavily contaminated with normal tissue, digestion products for both alleles will be obtained in a ratio that would give a false‐negative result.
	Depending on the regions analysed, it is possible that this technique may detect focal rather than whole arm deletions.
	Imbalanced polysomy, e.g. gain of 1 copy of chromosome 1 and 19, may result in allelic imbalance and be interpreted as loss of heterozygosity.
Comparative quantitative PCR	PCR cannot distinguish between absolute deletion and relative deletions in the presence of polyploidy (i.e. those deletions that would give a 2:4 ratio/equivalent with FISH).	If tumour samples are heavily contaminated with normal tissue the amount of PCR product obtained would result in a false‐negative result.
	Polysomy which causes the PCR product from control regions to increase could result in false‐positive results.	False‐negative results could be obtained if there has been an LOH without copy number reduction.
		Aneuploidy which causes the PCR product from control regions to decrease could result in false‐negative results.
MLPA	Cannot distinguish between absolute deletion and relative deletions in the presence of polyploidy (i.e. those deletions that would give a 2:4 ratio/equivalent with FISH).	If tumour samples are heavily contaminated with normal tissue, a false‐negative result may arise.
	SNPs at primer binding sites, as single mismatches at ligation sites can inhibit ligation.	False‐negative results could be obtained if there has been an LOH without copy number reduction.
CGH	Cannot distinguish between absolute deletion and relative deletions in the presence of polyploidy (i.e. those deletions that would give a 2:4 ratio/equivalent with FISH).	If tumour samples are heavily contaminated with normal tissue, a false‐negative result may arise.
		False‐negative results could be obtained if there has been an LOH without copy number reduction.
aCGH	As for CGH.	As for CGH.
SNP arrays	Cannot distinguish between absolute deletion and relative deletions in the presence of polyploidy arising from whole genome duplication after the codeletion event (i.e. those deletions that would give a 2:4 ratio/equivalent with FISH).	If tumour samples are heavily contaminated with normal tissue, a false‐negative result may arise.
Methylation arrays	Cannot distinguish between absolute deletion and relative deletions in the presence of polyploidy arising from whole genome duplication after the codeletion event (i.e. those deletions that would give a 2:4 ratio/equivalent with FISH).	If tumour samples are heavily contaminated with normal tissue, a false‐negative result may arise.
		False‐negative results could be obtained if there has been an LOH without copy number reduction.
NGS	Cannot distinguish between absolute deletion and relative deletions in the presence of polyploidy arising from whole genome duplication after the codeletion event (i.e. those deletions that would give a 2:4 ratio/equivalent with FISH).	If tumour samples are heavily contaminated with normal tissue, a false‐negative result may arise.

aCGH: array comparative genomic hybridisation; CGH: comparative genomic hybridisation; CISH: chromogenic in situ hybridisation; FFPE: formalin‐fixed, paraffin‐embedded tissue; FISH: fluorescence in situ hybridisation; LOH: loss of heterozygosity; MLPA: multiplex‐ligation‐dependent probe amplification; NGS: next‐generation sequencing; PCR: polymerase chain reaction; RFLP: restriction fragment length polymorphism; SNP: single nucleotide polymorphism.