OTUX provides a set of databases for various v-regions or their stretches. As we know, in 16S rRNAS research, different studies may target different v-regions. So, to enable cross-comparison of results obtained from OTUX, we have created a mapping back matrix. Using this matrix one can map-back the results obtained in terms of OTUX OTU IDs to Greengenes OTU IDs, which is one of the widely used reference database for OTU picking.

The OTUX database was created using the ‘prokMSA’ unaligned sequences from Greengenes database (v13.8) along with their taxonomic classification as well as Greengenes OTU IDs. We extracted the V regions from each sequence present in the database. The extracted sequences were then clustered based on sequence similarity using CD-HIT, wherein each resultant cluster constitutes sequences which shared 99% sequence identity with each other. Each cluster was assigned a v-region specific OTUX_V ID and was compiled to constitute a OTUX_V reference database for that v-region or a stretch of v-regions. The following procedure is explained using V4 region as an example. After compiling the OTUX databases, we calculated the propensity of a OTUX_V4 ID (OTUX_V4i) to be associated to a Greengenes OTU ID (GG_j) using the following formula,

The propensity matrix MAPMAT_V4, for the OTUX_V4 database, is then populated by computing all values for MAPMAT_V4ij where,

• i = 1 to N_OTUX, where N_OTUX is the total number of OTUXV4 OTUs
• j = 1 to N_GG, where N_GG is the total number of Greengenes OTUs

Now we have a mapping back matrix MAPMAT_V4, which is a N_GG × N_OTUX matrix. This matrix can now be used to map-back the classification results obtained in terms of OTUX_V4 OTUs to represent the results in terms of Greengenes (v13.8) OTU IDs. This mapping back can be either be ‘one-to-one mapping’ or ‘one-to-many mapping’ depending on the user criteria.

In this procedure each of the query sequences is assigned to particular Greengenes OTU IDs. For a particular query sequence ‘s’ which has been assigned to the OTU OTUX_V4x, the MAPMAT_V4 elements {MAPMAT_V4xj} are retrieved (wherein ‘j’ = 1 to NGG i.e. the total no. of Greengenes OTUs).The maximum value of {MAPMAT_V4xj} is computed. The sequence ‘s’ is classified to the Greengenes OTU ‘y’ (GG_y), wherein MAPMAT_V4xy = max{MAPMAT_V4xj}. The process is repeated for all query sequences and subsequently an OTU abundance table (T_GG), in terms of Greengenes OTU IDs, is generated by cumulating the total number of sequenced reads from the given metagenomic sample that could be classified/ attributed to each of the Greengenes OTUs.

This mapping is used to represent the abundance profile of a microbial community structure pertaining to a given metagenomic sample, wherein the abundance of each microbe (OTU) is represented in terms percentage normalized values. To begin with, for a set of query sequences corresponding to a metagenomic sample, the abundance table/profile T_OTUX is generated wherein the total number of sequences assigned to each of the OTUX_V4 OTUs are represented. For example, T_OTUX can be represented in form of a column matrix (of size NOTUX × 1) as depicted above wherein ‘i’ varies from 1 N_OTUX , i.e. the total number of OTUX_V4 OTUs, and wherein ‘a’ is the number of sequences assigned to the OTU OTUX_V41, ‘b’ is the number of sequences assigned to OTUX_V42, ‘c’ is the number of sequences assigned to OTUX_V43, and so on. Next, an OTU abundance table/profile (T_GGraw) is obtained for the set of query sequences, in terms of Greeengenes OTU IDs by multiplying the matrix MAPMAT_V4 with the matrix T_OTUX. It may be noted that given the nature of the MAPMAT matrix, the abundance values for each of the Greengenes OTUs in T_GGraw may be a fractional value.

T_GGraw = MAPMAT_V4 × T_OTUX

Wherein, T_GGraw is a column matrix of size (N_GG × 1), and N_GG is the total number of Greengenes OTUs. Finally a percentage normalized OTU abundance table/profile (T_GG%) is obtained by performing the following transformation on each element of T_GGraw

Wherein, T_GG% is a column matrix of size (NGG×1), and N_GG is the total number of Greengenes OTUs. The abundance of taxonomic groups present in the metagenomic sample, as obtained in the form of either of the three column matrices, viz. T_OTUX, T_GG and T_GG%, are further represented at any desired taxonomic level utilizing the taxonomic hierarchy information associated with the Greengenes OTUs.