Dataset Information

Deep learning-based selection of human sperm with high DNA integrity.

ABSTRACT: Despite the importance of sperm DNA to human reproduction, currently no method exists to assess individual sperm DNA quality prior to clinical selection. Traditionally, skilled clinicians select sperm based on a variety of morphological and motility criteria, but without direct knowledge of their DNA cargo. Here, we show how a deep convolutional neural network can be trained on a collection of ~1000 sperm cells of known DNA quality, to predict DNA quality from brightfield images alone. Our results demonstrate moderate correlation (bivariate correlation ~0.43) between a sperm cell image and DNA quality and the ability to identify higher DNA integrity cells relative to the median. This deep learning selection process is directly compatible with current, manual microscopy-based sperm selection and could assist clinicians, by providing rapid DNA quality predictions (under 10 ms per cell) and sperm selection within the 86th percentile from a given sample.

SUBMITTER: McCallum C

PROVIDER: S-EPMC6610103 | biostudies-other | 2019

REPOSITORIES: biostudies-other

ACCESS DATA

Publications

Deep learning-based selection of human sperm with high DNA integrity.

McCallum Christopher C Riordon Jason J Wang Yihe Y Kong Tian T You Jae Bem JB Sanner Scott S Lagunov Alexander A Hannam Thomas G TG Jarvi Keith K Sinton David D

Communications biology 20190703

Despite the importance of sperm DNA to human reproduction, currently no method exists to assess individual sperm DNA quality prior to clinical selection. Traditionally, skilled clinicians select sperm based on a variety of morphological and motility criteria, but without direct knowledge of their DNA cargo. Here, we show how a deep convolutional neural network can be trained on a collection of ~1000 sperm cells of known DNA quality, to predict DNA quality from brightfield images alone. Our resul ...[more]

PMID: 31286067

Similar Datasets

Project description:BackgroundProtamination and condensation of sperm chromatin as well as DNA integrity play an essential role during fertilization and embryo development. In some mammals, like pigs, ejaculates are emitted in three separate fractions: pre-sperm, sperm-rich (SRF) and post sperm-rich (PSRF). These fractions are known to vary in volume, sperm concentration and quality, as well as in the origin and composition of seminal plasma (SP), with differences being also observed within the SRF one. Yet, whether disparities in the DNA integrity and chromatin condensation and protamination of their sperm exist has not been interrogated.ResultsThis study determined chromatin protamination (Chromomycin A3 test, CMA3), condensation (Dibromobimane test, DBB), and DNA integrity (Comet assay) in the pig sperm contained in the first 10 mL of the SRF (SRF-P1), the remaining portion of the sperm-rich fraction (SRF-P2), and the post sperm-rich fraction (PSRF). While chromatin protamination was found to be similar between the different ejaculate fractions (P > 0.05), chromatin condensation was seen to be greater in SRF-P1 and SRF-P2 than in the PSRF (P = 0.018 and P = 0.004, respectively). Regarding DNA integrity, no differences between fractions were observed (P > 0.05). As the SRF-P1 has the highest sperm concentration and ejaculate fractions are known to differ in antioxidant composition, the oxidative stress index (OSi) in SP, calculated as total oxidant activity divided by total antioxidant capacity, was tested and confirmed to be higher in the SRF-P1 than in SRF-P2 and PSRF (0.42 ± 0.06 vs. 0.23 ± 0.09 and 0.08 ± 0.00, respectively; P < 0.01); this index, in addition, was observed to be correlated to the sperm concentration of each fraction (Rs = 0.973; P < 0.001).ConclusionWhile sperm DNA integrity was not found to differ between ejaculate fractions, SRF-P1 and SRF-P2 were observed to exhibit greater chromatin condensation than the PSRF. This could be related to the OSi of each fraction.

Dataset Information

Deep learning-based selection of human sperm with high DNA integrity.

Publications

Deep learning-based selection of human sperm with high DNA integrity.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets