Fig 1.
Workflow diagram outlining the process for both versions of the SGSA and the differences in processioning time between them.
The workflow for the SGSA v. 2 consists of multiple small improvements, which increase the throughput of the assay and reduce the hands-on time required to process 96 samples. Colony PCR saves four hours of labor compared to DNA extraction on a 96 well plate, and up to 1.5 days considering time saved from not having to grow a broth culture from colonies.
Fig 2.
SGSA v. 2 array images processed by the Alere ArrayMate for the reference Salmonella ser.
Typhimurium (1,4[5],12:i:-). A DNA stock was serially diluted and used as input into the SGSA v. 2 and imaged using the Alere ArrayMate system. The processed image files were used to create probe signal intensity files, which were used as input to the array classifier. Probe signals are consistent from the stock to a 10−3 dilution factor. Probe signals begin to be lost at 10−4 dilution and no signals are observed at 10−5 dilution. The limit of detection was set at 10−3 since it was the lowest dilution to completely identify the sample.
Table 1.
Draft genome assemblies were queried for the presence of the SGSA v. 2 probe sequences and the sequence identities were converted into simulated probe signal values.
These simulated values were processed using the array classifier and the antigenic formula and serotype were compared to the reported serotype for the assembly. Records were assigned to five different categories as described in the materials and methods.
Table 2.
In silico performance of the SGSA v. 2 classifications based on simulated probe intensities for serotypes with at least 100 samples tested. Records were assigned to five different categories as described in the materials and methods.
Table 3.
A panel of 406 isolates were analyzed using the SGSA v. 2 to determine the performance of the wet-lab assay.
Serotype predictions were compared to the phenotypically reported serotype and grouped into one of six categories.