Removing Noise From Pyrosequenced Amplicons

BMC Bioinformatics

Table 1 The eight GS FLX and one Titanium test data sets used in this study

Name	Type	Abundances	16S region	Read no.	Filtered no.	DeNoiser no.
Divergent	23 clones	Even	V5	57,902	35,190	42,052
Artificial	90 clones	Uneven	V5	46,249	31,867	37,903
Mock Communities
Even1	67 genomic	Even	V2	63,780	53,771	55,398
Even2	DNA isolates	Even	V2	53,763	45,178	46,294
Even3	DNA isolates	Even	V2	67,182	54,153	55,797
Uneven1	DNA isolates	Uneven	V2	54,099	44,926	46,837
Uneven2	DNA isolates	Uneven	V2	51,439	44,176	44,880
Uneven3	DNA isolates	Uneven	V2	60,976	50,931	53,225
Titanium	91 clones	Even	V4 - V5	62,873	25,438	21,477

The 16S rRNA region amplified, and whether references were mixed in equal, 'Even', or varying, 'Uneven' proportions are summarised together with the read number, following filtering and after the QIIME filtering used by the DeNoiser algorithm. All data sets were generated by GS FLX except that denoted Titanium.

ISSN: 1471-2105