How GC bias can skew species quantification and how to help stop it.

Updated: Dec 13, 2018

As if we didn’t have enough to worry about when analyzing microbiome samples, GC bias can come in misrepresent abundance data. Simply put, GC bias is when different bacterial species are sequenced together, and the species which is rich in GC content doesn’t get amplified as well because polymerases can have trouble reading over GC content. When the polymerase slows or pauses at (or sometimes doesn’t even gain access to) the GC regions, it can lead to that species looking like it has lower abundance in a community than it actually does.

As we all know, the more accurate quantification we can get for samples the better. Especially when testing samples from different environments (and therefore a higher chance to encounter species with variable GC content), there’s a clear need for a kit and protocol that’s flexible for the unknown. We don’t want to miss something important just because it has lots of GC content.

Thankfully, at Loop we obsess about GC bias and how to fix it. Of course there are DIY solutions - optimizing buffer conditions, adjusting PCR temperatures, or even engineering new polymerases -, but to keep things easy for you, we’ve done the hard part. We have identified and tested an ideal combination of buffer, temperature, and polymerase that minimizes GC bias.

But let’s show, not tell. To demonstrate the Loop kit’s ability to accurately measure relative species abundance in a microbial community containing species with different GC content, we used LoopSeq 16S Microbiome kit to analyze an ATCC 20 Strain Staggered Mix Genomic Material standard (Cat.# MSA-1003, figure below).

Figure 1: Demonstrates consistent results across species between the measured and expected abundances despite large amounts of GC content.

Comparison of the ten most abundant microbes in the sample show close agreement of measured abundance to expected. In addition, we see no GC bias, as shown in the close agreement of measured (red bars) versus expected (gray bars) across species with different levels of GC content (blue line). We’re pretty proud of that.

We never lose sight of the fact that a study is only as good as the data. For more accurate results, we strongly encourage thinking about accounting for GC bias, and we would be delighted if you choose a Loop kit to help. Learn more about Loop’s products at How It Works.