Forest Pathology and Mycology Laboratory, Department of Environmental Science, Policy and Management, University of California, Berkeley, California, United States of America.
PLoS One. 2013 Apr 30;8(4):e62419. doi: 10.1371/journal.pone.0062419. Print 2013.
Despite recent advances spearheaded by molecular approaches and novel technologies, species description and DNA sequence information are significantly lagging for fungi compared to many other groups of organisms. Large scale sequencing of vouchered herbarium material can aid in closing this gap. Here, we describe an effort to obtain broad ITS sequence coverage of the approximately 6000 macrofungal-species-rich herbarium of the Museum of Natural History in Venice, Italy. Our goals were to investigate issues related to large sequencing projects, develop heuristic methods for assessing the overall performance of such a project, and evaluate the prospects of such efforts to reduce the current gap in fungal biodiversity knowledge. The effort generated 1107 sequences submitted to GenBank, including 416 previously unrepresented taxa and 398 sequences exhibiting a best BLAST match to an unidentified environmental sequence. Specimen age and taxon affected sequencing success, and subsequent work on failed specimens showed that an ITS1 mini-barcode greatly increased sequencing success without greatly reducing the discriminating power of the barcode. Similarity comparisons and nonmetric multidimensional scaling ordinations based on pairwise distance matrices proved to be useful heuristic tools for validating the overall accuracy of specimen identifications, flagging potential misidentifications, and identifying taxa in need of additional species-level revision. Comparison of within- and among-species nucleotide variation showed a strong increase in species discriminating power at 1-2% dissimilarity, and identified potential barcoding issues (same sequence for different species and vice-versa). All sequences are linked to a vouchered specimen, and results from this study have already prompted revisions of species-sequence assignments in several taxa.
尽管最近在分子方法和新技术的推动下取得了进展，但与许多其他生物群体相比，真菌的物种描述和 DNA 序列信息仍然严重滞后。对有凭证标本的大规模测序可以帮助缩小这一差距。在这里，我们描述了一项努力，即在意大利威尼斯自然历史博物馆拥有约 6000 种丰富大型真菌标本的标本馆中获得广泛的 ITS 序列覆盖。我们的目标是研究与大型测序项目相关的问题，开发评估此类项目整体性能的启发式方法，并评估此类努力减少真菌生物多样性知识当前差距的前景。这项工作共生成了 1107 条提交给 GenBank 的序列，其中包括 416 个以前未代表的分类群和 398 个与未识别环境序列最佳 BLAST 匹配的序列。标本的年龄和分类群影响测序成功率，对失败标本的后续工作表明，ITS1 迷你条形码大大提高了测序成功率，而不会大大降低条形码的区分能力。基于成对距离矩阵的相似性比较和非度量多维标度排序被证明是验证标本鉴定整体准确性、标记潜在错误鉴定和识别需要进一步物种水平修订的分类群的有用启发式工具。种内和种间核苷酸变异的比较表明，在 1-2%的差异下，物种区分能力有了很大的提高，并确定了潜在的条形码问题（不同物种的相同序列和反之亦然）。所有序列都与有凭证的标本相关联，这项研究的结果已经促使对几个分类群的物种-序列分配进行了修订。
