BRAF V600 variants, especially BRAF V600E, are known to be important driver mutations in melanoma (1,2), nonsmall cell lung cancer (3), and colon cancer (4), as BRAF inhibitors ± MEK inhibitors have been approved for treatment in these cancers. Thus, finding BRAF V600 variants in these cancers is mandatory to provide optimal treatment. In the perfect world, next generation sequencing (NGS) of all cancers would be possible at diagnosis. However, this is not always the case, so robust, alternative methods of detection are welcomed.In the present issue of Translational Cancer Research, Kang et al. build a prediction model to detect BRAF V600 variants with mRNA gene expression data in various cancer types. The authors obtained mRNA gene expression data of BRAF V600-variant cancers from The Cancer Genome Atlas (TCGA) pan-cancer database, and constructed a training set from thyroid carcinoma, cutaneous melanoma and colon adenocarcinoma cases, which are known to have high prevalence of BRAF V600 alterations. The authors then adopted a penalized logistic regression for prediction of BRAF V600E variants. Area under the receiver operating characteristic (AUROC) and area under the precisionrecall (AUPR) for the test set was 0.98 and 0.98 in thyroid carcinoma, 0.90 and 0.71 in colon carcinoma, and 0.85 and 0.65 in cutaneous melanoma, respectively. However, AUROC and AUPR was low in the unseen test set for cancer types with low prevalence of BRAF V600 variants. These results suggested that this prediction model can reliably detect BRAF V600E-variant cases using mRNA expression data in cancer types with high incidence of BRAF V600E, but not those with low incidence.