Send to

Choose Destination
Biomed Chromatogr. 2017 Aug;31(8). doi: 10.1002/bmc.3931. Epub 2017 Feb 5.

Untargeted metabolomic profiling of seminal plasma in nonobstructive azoospermia men: A noninvasive detection of spermatogenesis.

Author information

Reproductive Biotechnology Research Center, Avicenna Research Institute, ACECR, Tehran, Iran.
Department of Chemistry, Faculty of Sciences, Chemometrics Laboratory, Tarbiat Modares University, Tehran, Iran.
Nanobiotechnology Research Center, Avicenna Research Institute, ACECR, Tehran, Iran.
Department of Biotechnology, The Faculty of Renewable Emergies and New Technologies, Tehran, Iran.
Department of Nanobiotechnology, Protein Research Institute, Shahid Beheshti Universtiy, Tehran, Iran.
Department of Phytochemistry, Medicinal Plants and Drugs Research Institute, Shahid Beheshti University, Tehran, Iran.


Male factor infertility is involved in almost half of all infertile couples. Lack of the ejaculated sperm owing to testicular malfunction has been reported in 6-10% of infertile men, a condition named nonobstructive azoospermia (NOA). In this study, we investigated untargeted metabolomic profiling of the seminal plasma in NOA men using gas chromatography-mass spectrometry and advance chemometrics. In this regard, the seminal plasma fluids of 11 NOA men with TESE-negative, nine NOA men with TESE-positive and 10 fertile healthy men (as a control group) were collected. Quadratic discriminate analysis (QDA) technique was implemented on total ion chromatograms (TICs) for identification of discriminatory retention times. We developed multivariate classification models using the QDA technique. Our results revealed that the developed QDA models could predict the classes of samples using their TIC data. The receiver operating characteristic curves for these models were >0.88. After recognition of discriminatory retention time's asymmetric penalized least square, evolving factor analysis, correlation optimized warping and alternating least squares strategies were applied for preprocessing and deconvolution of the overlapped chromatographic peaks. We could identify 36 discriminatory metabolites. These metabolites may be considered discriminatory biomarkers for different groups in NOA.


GC-MS; chemometrics; male infertility; nonobstructive azoospermia; untargeted metabolomic profiling

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Wiley
Loading ...
Support Center