文章：An improved dual-indexing approach formultiplexed 16S rRNA gene sequencing on theIllumina MiSeq platform

   Illumina的Miseq和Hiseq机器的数据产生通量逐步增加，16s中每个样品所需的数据量较少，为此进行多样品的pooling测序，这个过程中产生了single-index和dual-index的测序，single-index测序需要大量的barcode序列，dual-index则减少了对barcode序列的需求，但是却没有解决16s序列复杂度较低的问题。

   在16s的研究中，对于测序来讲一个很大的问题是“low sequence diversity”，序列复杂度太低，这个不利于Illumina测序时的碱基信号校准，MiSeq (cluster identificationand phasing/prephasing rate determination) require abalanced base composition through the initial 12 to18 cycles of the run. 为此在文库中会需要加入Phix序列，一般来讲会是1：1添加，如此产生的序列Q30可达到70%。

   在本问题提出如下方案，加入了一段heterogeneity spacer序列增加文库的复杂性，以提升碱基质量。

   v3-v4区域大约469bp（a region of approximately 469 bp encompassing the V3 and V4 hypervariable regions of the16S rRNA gene was targeted for sequencing. ）整个序列包括：1) a linker sequence allowing amplicons to bind tothe flow cell and be sequenced using the standard IlluminaHP10 or HP11 (Illumina, San Diego, CA, USA) sequen-cing primers;  2) a 12 bp index sequence; 3) a 0 to 7 bp“heterogeneity spacer” that we designed in this studyto mitigate the issues caused by low sequence diversity amplicons，4) 16SrRNA gene universal primers。（在文章http://microbiomejournal.biomedcentral.com/articles/10.1186/2049-2618-2-6 中有附件可以查看）

   利用此方法进行建库，基本信息如下：Sequal-Prep Normalization Kit、11 pools with 271 to 426 samples perpool、AMPure XT beads、LabChip GX、Quantification Kit、PhiX Control library (v3)(Illumina) was combined with the amplicon library (expected at 20%) 、The library was clustered to a densityof approximately 570 K/mm2 、libraries were sequenced either on 250PE or 300PE MiSeq runs.

   文中数据处理的策略如下：1) removal of primer sequence; 2) truncation ofsequence reads not having an average quality of 20 over a 30 bp sliding window based on the phred algorithm implemented previously;  3) removal of trimmed reads having less than 75% of their original length, as wellas its paired read.  Further sequence reads processing was performed usingQIIME (version 1.6.0,) and included additional qualitytrimming, demultiplexing, and taxonomic assignments. QIIME quality trimming was performed using the followingcriteria: 1) truncate sequence reads before three consecutive low-quality bases and re-evaluate for length; 2) no ambiguous base calls; and 3) minimum sequence length of 150 bpafter trimming. 4）Paired-end reads were aligned to pre-aligned Greengene 16S rRNA gene sequences.

   利用此方法得到的数据质量如下：

   从上表看出，数据产量略低，数据质量比较理想，phix可以降低到较小的比例。

   对于v3-v4区域来讲，That said, the data-set generated with the 300PE protocol has an approximate90 bp overlap, making this protocol a preferred and super-ior approach that generates high-confidence paired-endassemblies compared to the 250PE, which at most overlapby 30 bp.