Nucleic Acids Research, 2018, Vol. 46, No. 20, Pages 10546–10562, https://doi.org/10.1093/nar/gky889
Our paper included a benchmark comparing nine methods for multi-omics clustering, including PINS (1). We reported that PINS crashed when run on a breast cancer dataset. We performed all the benchmarks on a 64-bit computer, using the 32-bit version of R. In later tests we observed that PINS did not crash on 64-bit R, and it only crashed on 32-bit R due to insufficient memory. The clustering that PINS obtained on the breast cancer dataset had 4 enriched clinical parameters, and the P-value for the logrank test on that clustering was 0.05.
REFERENCES
- 1. Nguyen T., Tagett R., Diaz D., Draghici S.. A novel approach for data integration and disease subtyping. Genome Res. 2017; 27:2025–2039. [DOI] [PMC free article] [PubMed] [Google Scholar]