BioCoder #9

Chapter 5. Standards for Protocols: The Quickest Way to Reproducibility

Ben Miles and Tali Herzka

Anyone who has worked in a lab is familiar with protocols that simply do not work as they should. Protocols are often subjective, written for oneself to use again in the future in the same lab. However, problems inevitably arise for the next scientist, whether in the same or a different lab, attempting to replicate the experiment that’s missing some critical details about how exactly to do that tricky incubation step. Looking at the literature today, there are many examples of methods sections that are overly edited and ambiguous. The loss of precise descriptions of methods most likely occurs out of necessity to edit articles to an appropriate length for the publication; however, precision and reproducibility should not be sacrificed for better word counts. Table 5-1 illustrates a number of typical examples, taken from the literature, of vague descriptions of processes that are open to interpretation. This lack of precision in documentation hasn’t gone unnoticed in the wider scientific community and is recognized as one of the causes of the widespread irreproducibility of scientific studies that is often discussed in the media. Thankfully, a solution to this problem is emerging in the form of standardized protocols.

Standard specifications can be used for describing protocols to eliminate their subjectivity and, as a consequence, variability. Picture a world where every protocol written ...