SCISIPBIO: A data-science approach to evaluating the likelihood of fraud and error in published studies

Biomedical research, including human genetics, is one of the largest areas of present-day science. A cause for concern is the fact that most research on human genes concentrates on only approximately 2,000 of the 19,000 human genes. We recently demonstrated that the vast differences in attention human genes receive can be accurately explained by a small set of identifiable properties of genes that make them easy to study and historical research patterns. Concerningly, we demonstrated that even highly promising genes that could already be studied by current technologies remain ignored. This suggests a great degree of inefficiency in the manner in which the biomedical research enterprise explores the biological unknown. In order to address this matter systematically and to provide guidance on what human genes may deserve more attention, we propose to create a self-updating monitoring system of the worldwide biomedical research enterprise on human genes.