Multiple sclerosis (MS) is a chronic, progressive, and debilitating disease of the central nervous system. This disease is thought to have multiple predisposing factors, including genetic risk factors, vitamin D deficiency, and infectious agents. Although a fungal presence in MS has long been suspected, only recently did studies associate the fungal gut microbiome—the mycobiome—with MS in adult individuals. However, investigations of the mycobiome’s association with disease have been hampered by a lack of standardized methodologies. Both the methods for sequencing and bioinformatic characterization of the mycobiome require systematic assessment in order to standardize such methodologies. Furthermore, although alterations in the mycobiome have been associated with MS in adults, no studies to date have examined if the same case can be made for pediatric-onset MS (POMS) individuals. We set out to evaluate the performance of different internal transcribed spacer (ITS) amplicon sequencing and bioinformatics approaches by varying the PhiX concentration and assessing the abilities of different analytical pipelines to characterize a predefined fungal community. We also used the most optimal methods resulting from our evaluation to examine the gut mycobiome in persons with and without POMS. Our methodology evaluation finds that a 50% PhiX spike-in increased the sequence quality, with 15% more bases at a Phred score of 30 or greater (≥ 99.9% accuracy) but decreased the overall number of reads by more than half when compared to a 25% PhiX spike-in. Of the three fungal metabarcoding bioinformatics pipelines we examined, LotuS classified the highest number of expected species. Although we did not find a significant difference in the fungal community between POMS and unaffected control individuals, we found that the genus Agaricus was highly abundant in POMS individuals; we also found an association of the species Candida albicans with POMS.