“Sharing GWAS Summary Statistics Results in More Citations”, Guillermo Reales, Chris Wallace2023-01-28 ()⁠:

Rates of sharing of genome-wide association studies (GWAS) summary statistics are historically low, limiting potential for scientific discovery.

Here we show, using GWAS Catalog data, that GWAS papers that share data get on average 81.8% more citations, an effect that is sustained over time…By adding a binary variable describing sharing practice, we concluded that sharing summary statistics has a positive effect on the RCR, providing ~81.8% more citations on average than non-sharing articles.

Figure 2: Citation patterns over time (2006–152021), measured in log relative citation ratio. (a) All GWAS. (b) Split by summary statistics sharing status. Sharing studies are consistently more cited than non-sharing studies. Lower and upper box hinges represent the 25<sup>th</sup> and 75<sup>th</sup> percentiles, respectively. The whiskers extend for 1.5 × IQR from each hinge, and the horizontal line within the boxes represents the median.
Figure 2: Citation patterns over time (2006152021), measured in log relative citation ratio. (a) All GWAS. (b) Split by summary statistics sharing status. Sharing studies are consistently more cited than non-sharing studies. Lower and upper box hinges represent the 25th and 75th percentiles, respectively. The whiskers extend for 1.5 × IQR from each hinge, and the horizontal line within the boxes represents the median.
Figure 3: Mean citation count evolution after publication, by year of publication (2010–82018). Sharing studies get more citations from early on, then stabilizing circa 2 years after publication. (a) Mean citation count ratio (shared/unshared). (b) Sharing (orange) and non-sharing (blue) mean citation count. Text in squares indicates the number of studies in each category.