The novel coronavirus disease (COVID-19) pandemic is emerging as a global health threat and shows a higher risk for men than women. Thus far, the studies on andrological consequences of COVID-19 are limited. To ascertain the consequences of COVID-19 on sperm parameters after recovery, we recruited 41 reproductive-aged male patients who had recovered from COVID-19, and analyzed their semen parameters and serum sex hormones at a median time of 56 days after hospital discharge. For longitudinal analysis, a second sampling was obtained from 22 of the 41 patients after a median time interval of 29 days from first sampling. Compared with controls who had not suffered from COVID-19, the total sperm count, sperm concentration, and percentages of motile and progressively motile spermatozoa in the patients were significantly lower at first sampling, while sperm vitality and morphology were not affected. The total sperm count, sperm concentration, and number of motile spermatozoa per ejaculate were significantly increased and the percentage of morphologically abnormal sperm was reduced at the second sampling compared with those at first in the 22 patients examined. Though there were higher prolactin and lower progesterone levels in patients at first sampling than those in controls, no significant alterations were detected for any sex hormones examined over time following COVID-19 recovery in the 22 patients. Although it should be interpreted carefully, these findings indicate an adverse but potentially reversible consequence of COVID-19 on sperm quality.

Chromosomal polymorphism has been reported to be associated with infertility, but its effect on IVF/ICSI-ET outcome is still controversial. To evaluate whether or not chromosomal polymorphism in men plays a role in spermatogenesis and the outcome of IVF/ICSI-ET, we retrospectively analysed 281 infertile couples. Measures included fertilization rate, implantation rate, pregnancy rate, clinical pregnancy rate, ongoing pregnancy rate, early miscarriage rate and preterm rate. Men with chromosomal polymorphism had significantly higher frequencies of severe oligozoospermia and azoospermia than those without (37.12% vs. 16.11%, p < 0.001; 27.27% vs. 10.74%, p < 0.001; respectively). Significantly, lower fertilization rate (68.02% vs. 78.00%, p < 0.001) and clinical pregnancy rate (45.00% vs. 66.67%, p = 0.031) were observed in polymorphism-carrying men with severe oligozoospermia compared with non-carriers with severe oligozoospermia. This suggests that chromosomal polymorphism has adverse effects on spermatogenesis, negatively influencing the outcome of IVF/ICSI-ET treatment. Polymorphic variations on the Y chromosome have been found to be the most prevalent polymorphism in infertile men, most frequently occurring in patients with severe oligozoospermia.

Aiming to represent user characteristics and personal interests, the task of user profiling is playing an increasingly important role for many real-world applications, e.g., e-commerce and social networks platforms. By exploiting the data like texts and user behaviors, most existing solutions address user profiling as a classification task, where each user is formulated as an individual data instance. Nevertheless, a user's profile is not only reflected from her/his affiliated data, but also can be inferred from other users, e.g., the users that have similar co-purchase behaviors in e-commerce, the friends in social networks, etc. In this paper, we approach user profiling in a semi-supervised manner, developing a generic solution based on heterogeneous graph learning. On the graph, nodes represent the entities of interest (e.g., users, items, attributes of items, etc.), and edges represent the interactions between entities. Our heterogeneous graph attention networks (HGAT) method learns the representation for each entity by accounting for the graph structure, and exploits the attention mechanism to discriminate the importance of each neighbor entity. Through such a learning scheme, HGAT can leverage both unsupervised information and limited labels of users to build the predictor. Extensive experiments on a real-world e-commerce dataset verify the effectiveness and rationality of our HGAT for user profiling.