Abstract
Purpose
Nucleolus-precursor body (NPB) migration speed in human zygotes predicts live birth after assisted reproduction. However, the impact of changes in NPB migration speed on live births was unclear. This study investigated the relationship between patterns of change in NPB migration speed over time and clinical outcomes.
Methods
The central coordinates of NPBs were tracked at multiple time points to assess changes in NPB migration speed from 1 to 7 h before the disappearance of the pronucleus. Time-series clustering was performed. Cluster centroids were calculated using the time-series K-means method. A training dataset (N = 210 NPBs) was used to define clusters and determine the optimal number. A test dataset (N = 141 NPBs) was then examined.
Results
When examining the live-birth rate in training data, the standardized residuals of Cluster 7 were significantly higher (100.0%, P < 0.01), while those of Cluster 2 were significantly lower (36.7%, P < 0.05). However, in the test data, the live-birth rate was not significantly different among the clusters. When we placed the clusters into three groups based on the live-birth rates in the training data (low, ≤ 50.0%; middle, 50.1–65.0%; high, > 65.0%), the cluster groups were correlated with live-birth rates in the test data. The live-birth rate in clusters comprising the high subgroup was significantly higher than that in the clusters comprising the low subgroup (80.0% vs. 47.4%, P = 0.034).
Conclusion
Time-series analysis of NPB migration speed may serve as a useful indicator for selecting human zygotes with a higher potential for viable pregnancies and live births following assisted reproduction.
Supplementary Information
The online version contains supplementary material available at 10.1007/s10815-025-03578-3.
Keywords: Live birth, Nucleolus-precursor body, Pronuclei, Time-series clustering, Cluster analysis, Zygote
Introduction
Nuclear bodies, membraneless organelles enriched with RNA and proteins, play an essential role in gene regulation [1]. In mammalian oocytes and zygotes, nucleolus-like bodies and nucleolus-precursor bodies (NPBs) are compact structures with inactive ribosome biosynthesis, distinct from the nucleoli found in somatic cells [2, 3]. During oocyte maturation, nucleolus-like bodies are present in fully grown, germinal-vesicle-stage oocytes and then disappear. Following meiotic division and fertilization, NPBs form in the male and female pronuclei (PN) of the zygote, contributing to genome reorganization and remodeling during early embryogenesis [3–5]. Maternal NPBs are not essential for oocyte maturation, but NPBs are indispensable for embryonic development following fertilization and may play an active role in the formation of centromeric chromatin [4, 6].
In murine zygotes, the NPBs are surrounded by pericentromeric and centromeric heterochromatin in a ring-like shape [5]. In human zygotes, NPB position correlates with chromatin distribution, and NPB repositioning reflects a global redistribution and clustering of chromatin at the interpronuclear interface [7, 8]. The parental genomes cluster around nucleoli within each pronucleus, and this clustering is essential for the successful merging of parental genomes after fertilization in human zygotes [8].
Morphological classification of zygotes prior to first cleavage has been widely applied as a tool for predicting subsequent embryonic development after assisted reproductive technology (ART) including classifications based on PN size, the orientation of the first mitotic spindle, and the number and distribution of NPBs in the PN [9–13]. Focusing on the zygote in the pronuclear stage, Tesarik and Greco [14] reported a simple and non-invasive method for predicting the developmental potential of human zygotes based on the distribution and number of NPBs in the PNs. Since their report, several studies have investigated the relationship between NPB localization and embryonic development, with some supporting the predictive value of these parameters [15–17]. However, these findings also suggest that more detailed analyses are needed to fully understand the role of these essential structures in development and to optimize their use in selecting high-quality embryos [18, 19].
The NPBs move extensively, and the characteristic patterns of this movement change significantly during syngamy [20, 21]. Previous studies have demonstrated that the NPB migration speed in human zygotes predicts the competence of the embryo to develop to the blastocyst stage and result in a live birth after intracytoplasmic sperm injection (ICSI) or in vitro fertilization (IVF) [22, 23]. However, it was unclear whether changes in the NPB migration speed over time were correlated with live birth after ART. Therefore, this study investigated the relationship between the time-series patterns of change in the NPB migration speed and clinical outcomes following ART.
Materials and methods
Datasets and patients
The datasets examined, which included the coordinates of NPBs and PNs, were used previously to explore the relationship between migration speed and clinical outcomes after ART [22, 23]. The present study, however, focused on a novel analysis of patterns in NPB migration speed over time and their association with clinical outcomes following ART. We combined three previously studied datasets to compile the dataset for this analysis. We included NPBs from both male and female PNs and NPBs from zygotes generated using either IVF or ICSI, because we aimed to obtain generalized data pertaining to NPB migration speed with the potential for clinical application. One dataset was from vitrified-warmed, single ICSI-derived blastocyst transfer cycles conducted between April 2018 and June 2019. The dataset included 30 patients with live births and 30 patients with no pregnancy (gestational sac negative). The mean age of the women at the time of ovum pick-up (OPU) was 34.2 ± 3.9 years (range 27–43 years). The second dataset was from 14 zygotes (4 patients with live birth and 10 patients with no pregnancy) that developed from ICSI-derived fertilized oocytes between January 2019 and May 2019 and underwent vitrified-warmed, single blastocyst transfer. The mean age of the women in this cohort was 34.3 ± 2.9 years (range 30–38 years) at the time of OPU. The third dataset was from patients who underwent IVF with vitrified-warmed, single IVF-derived blastocyst transfer cycles between October 2019 and September 2021. The mean age of the women in this group was 33.0 ± 3.2 years (range 28–40 years) at the time of OPU. This dataset included 19 patients with live births and 19 patients who did not achieve pregnancy. The combined dataset included balanced numbers of NPBs from IVF and ICSI (± 2% in the training data and ± 9% in the test data) and from male and female PNs (± 5% in both datasets). All zygotes with complete coordinate data for NPBs from 1 to 7 h before PN disappearance within the study period were included in the analysis.
This study was approved by our Institutional Review Board, and informed consent was obtained from all patients. All retrospective analyses were conducted with patient identities masked.
Time-series clustering of the migration speed of nucleolus-precursor bodies
NPB migration speed was measured using time‐lapse microscopy as described previously [22, 23]. When selecting NPBs for analysis, we first selected NPBs with the longest possible appearance period. Secondly, we chose NPBs that exhibited consistent movement throughout the observation period. The central coordinates of the PNs and NPBs were measured using Kinovea motion capture software (Patreon, San Francisco, CA, USA) and were confirmed or revised in every image. The migration distance of the NPBs between two sequential images was calculated relative to the central coordinates of the PN.
NPB positions change over time until PN breakdown. In order to ensure consistency, we performed time-series clustering analysis using NPB positions recorded from 1 to 7 h before PN disappearance, with the time of PN disappearance as the reference point. The analysis period was set because this was the time interval with the most NPB data available for comparison (using rate: 58.0%). The data obtained were randomly split into training and test datasets in a 60:40 ratio (210 NPBs:141 NPBs). To prevent overfitting and to evaluate the generalizability of the model, we split the dataset randomly to ensure that both the training and test datasets maintained a similar distribution and to minimize potential bias. Z-score standardization was applied based on the raw distance information to ensure that different data scales (i.e., differences in baseline speed) could be compared effectively. This transformation ensured that clustering was based on relative movement patterns rather than absolute speed differences and prevented baseline speed variations among groups from directly affecting the clustering results.
The number of clusters was determined by the elbow method and silhouette analysis on the training dataset (Fig. 1). Subsequently, time-series clustering was performed using the time-series K-means method. The cluster centroids were calculated by taking the average values at each time point for all time-series data within the cluster and connecting those averages to represent the central tendency of the cluster (Fig. 2). The test dataset was then categorized based on these cluster centroids.
Fig. 1.
Determining the optimal number of clusters. Elbow method to determine optimal number of clusters (a). Silhouette scores for different numbers of clusters (b)
Fig. 2.
Time-series clustering of migration speed of nucleolus-precursor bodies from 1 (the 60th frame) to 7 (the 0th frame) hours before pronuclear disappearance. Red: cluster centroid. Blue: migration speed of each nucleolus-precursor body (NPB) in a male pronucleus. Orange: migration speed of each NPB in a female pronucleus. Frames were captured at 5-min intervals. Clustering analysis identified distinct migration patterns summarized as follows. Cluster 0 (N = 25): exhibits a random migration pattern with no clear trend, showing only minor fluctuations throughout the time series. Cluster 1 (N = 18): characterized by consistently low migration speed with minimal variation, apart from a slight increase around frame 30 (4 h before pronuclei (PN) disappearance). Cluster 2 (N = 30): initially slow movement followed by a gradual increase in speed around frame 30 (4 h before PN disappearance), with moderate fluctuations throughout. Cluster 3 (N = 17): displays the least movement among all clusters, remaining largely stable except for a small peak between frames 20 and 40 (from 3 to 5 h before PN disappearance). Cluster 4 (N = 22): exhibits higher migration speed with periodic peaks around frames 20 and 40 (from 3 to 5 h before PN disappearance), suggesting dynamic fluctuations. Cluster 5 (N = 30): similar to Cluster 4 but with more pronounced rhythmic fluctuations, including two distinct peaks and a slight decrease after frame 50 (2 h before PN disappearance). Cluster 6 (N = 27): displays moderate migration speed variations with a peak around frame 30, followed by stable movement. Cluster 7 (N = 8): shows a distinct delayed response, with a sharp increase in migration speed after frame 50 (2 h before PN disappearance) following a prolonged slow phase. Cluster 8 (N = 33): primarily low migration speed, similar to Cluster 1, but with a brief increase around frames 30–40 (from 3 to 4 h before PN disappearance), followed by a decline
To evaluate whether well-characterized morphokinetic parameters assessed prior to the development of a 2-cell embryo were correlated with the time-series clusters, we quantified the time of the onset of first cleavage and the length of the first mitosis (the time between PN disappearance and the completion of division resulting in 2 cells).
The randomization of dataset, Z-score standardization, time-series clustering, the elbow method, and silhouette analysis were conducted using Python (version 3.12.3; https://www.python.org/) within the Jupyter Notebook (version 7.0.8; Project Jupyter, https://jupyter.org/) environment, with package management handled by the Anaconda Navigator (version 2.6.2; Anaconda, Inc, Austin, TX, USA).
Statistical analysis
All data are presented as mean ± standard deviation. Normality was assessed using the Shapiro–Wilk test. Accordingly, the Mann–Whitney U test was used to compare differences between two groups, and the Kruskal–Wallis test was applied to compare differences among more than two groups. The rates of live births were evaluated using Fisher’s exact probability test and residual analysis for each cluster (0–8) in both the training data and the test data because there were cells contained zero values in both the training and test data. The live-birth rates in the clusters, categorized as high, middle, or low, were evaluated using Holm’s test. The comparison of live-birth rates between the training and test data categorized by cluster was evaluated using Fisher’s exact probability test. A P-value of < 0.05 was considered statistically significant. Statistical analyses (with the exception of the residual analysis) were performed using EZR software [24]. Residual analysis was performed using js-STAR XR + (release 2.1.3 j, https://www.kisnet.or.jp/nappa/software/star/).
Results
Time-series clustering using the elbow method and silhouette scores identified nine clusters as the most appropriate number of clusters (silhouette score 0.060), and these were designated as Cluster 0–Cluster 8 (Fig. 1). To examine whether known heterogeneity in the combined dataset influenced the cluster analysis, we first examined whether differences in fertilization methods and the localization of NPBs in either the male or female PN affected NPB migration speed. The migration speed of NPBs in the male pronucleus (mNPBs) was significantly faster than that of NPBs in the female pronucleus (fNPBs) in both ICSI- and IVF-derived zygotes (ICSI: mNPBs, 3.88 ± 1.17 vs. fNPBs, 3.40 ± 0.98 µm/h, P = 0.014; IVF: mNPBs, 3.55 ± 1.14 vs. fNPBs, 3.23 ± 1.20 µm/h, P = 0.031; Fig. 3). Additionally, the migration speed of mNPBs was significantly faster in ICSI-derived zygotes than in IVF-derived zygotes (P = 0.021; Fig. 3). Although not statistically significant, the migration speed of fNPBs tended to be faster in ICSI-derived zygotes than in IVF-derived zygotes (P = 0.057; Fig. 3).
Fig. 3.
The migration speed of the nucleolus-precursor body (NPB) in both male and female pronuclei (mNPB and fNPB) and the insemination methods. ICSI-mNPB, N = 89; ICSI-fNPB, N = 77; IVF-mNPB, N = 91; IVF-fNPB, N = 94. Mann–Whitney U test
After standardizing for differences in baseline migration speed, we examined whether differences in fertilization methods or localization to either the male or female PN drove categorization of the NPBs into clusters in the training and test datasets. The fertilization method and type of PN where the NPB was localized did not differ significantly among the clusters (Tables 1 and 2).
Table 1.
Origin of the NPBs in the training dataset
| No. of NPBs | IVF n (%) |
ICSI n (%) |
P-valuea | mNPB n (%) |
fNPB n (%) |
P-valuea | |
|---|---|---|---|---|---|---|---|
| Cluster 0 | 25 | 13 (52.0) | 12 (48.0) | 0.741 | 18 (72.0) | 7 (28.0) | 0.678 |
| Cluster 1 | 18 | 11 (61.1) | 7 (38.9) | 9 (50.0) | 9 (50.0) | ||
| Cluster 2 | 30 | 16 (53.3) | 14 (46.7) | 17 (56.7) | 13 (43.3) | ||
| Cluster 3 | 17 | 9 (52.9) | 8 (47.1) | 11 (64.7) | 6 (35.3) | ||
| Cluster 4 | 22 | 11 (50.0) | 11 (50.0) | 10 (45.5) | 12 (54.5) | ||
| Cluster 5 | 30 | 15 (50.0) | 15 (50.0) | 14 (46.7) | 16 (53.3) | ||
| Cluster 6 | 27 | 9 (33.3) | 18 (66.7) | 11 (40.7) | 16 (59.3) | ||
| Cluster 7 | 8 | 5 (62.5) | 3 (37.5) | 5 (62.5) | 3 (37.5) | ||
| Cluster 8 | 33 | 19 (57.6) | 14 (42.4) | 15 (45.5) | 18 (54.5) |
fNPBs, nucleolus-precursor bodies in female pronuclei; ICSI, intracytoplasmic sperm injection; IVF, in vitro fertilization; mNPBs, nucleolus-precursor bodies in male pronuclei
aFisher’s exact test
Table 2.
Origin of NPBs in the test dataset
| No. of NPBs | IVF n (%) |
ICSI n (%) |
P-valuea | mNPB n (%) |
fNPB n (%) |
P-valuea | |
|---|---|---|---|---|---|---|---|
| Cluster 0 | 17 | 8 (47.1) | 9 (52.9) | 0.705 | 12 (70.6) | 5 (29.4) | 0.159 |
| Cluster 1 | 17 | 11 (64.7) | 6 (35.3) | 7 (41.2) | 10 (58.8) | ||
| Cluster 2 | 12 | 9 (75.0) | 3 (25.0) | 4 (33.3) | 8 (66.7) | ||
| Cluster 3 | 12 | 6 (50.0) | 6 (50.0) | 3 (25.0) | 9 (75.0) | ||
| Cluster 4 | 16 | 10 (62.5) | 6 (37.5) | 5 (31.3) | 11 (68.8) | ||
| Cluster 5 | 13 | 8 (61.5) | 5 (38.5) | 8 (61.5) | 5 (38.5) | ||
| Cluster 6 | 24 | 11 (45.8) | 13 (54.2) | 11 (45.8) | 13 (54.2) | ||
| Cluster 7 | 3 | 1 (33.3) | 2 (66.7) | 1 (33.3) | 2 (66.7) | ||
| Cluster 8 | 27 | 13 (48.1) | 14 (51.9) | 16 (59.3) | 11 (40.7) |
fNPBs, nucleolus-precursor bodies in female pronuclei; ICSI, intracytoplasmic sperm injection; IVF, in vitro fertilization; mNPBs, nucleolus-precursor bodies in male pronuclei
aFisher’s exact test
Assured that neither fertilization method nor NPB localization (male PN vs. female PN) influenced clustering, we continued to assess the characteristics and predictive value of the nine distinct clusters identified in the time-series clustering analysis of NPB migration speed patterns (Fig. 2). The time of onset of the first cleavage and the length of the first mitosis, two well-characterized morphokinetic parameters assessed prior to the development of a 2-cell embryo, were not significantly different among the clusters in either the training or test data (Tables 3 and 4).
Table 3.
The relationship between each cluster and morphokinetic parameters until the 2-cell embryonic stage in the training dataset
| The onset of the first cleavage (hours)a | P-valueb | The length of the first mitosis (hours)a,c | P-valueb | |
|---|---|---|---|---|
| Cluster 0 | 24.08 ± 2.43 | 0.863 | 2.53 ± 0.43 | 0.481 |
| Cluster 1 | 24.51 ± 2.47 | 2.37 ± 0.54 | ||
| Cluster 2 | 23.48 ± 2.27 | 2.45 ± 0.27 | ||
| Cluster 3 | 24.37 ± 2.80 | 2.40 ± 0.27 | ||
| Cluster 4 | 24.22 ± 2.44 | 2.38 ± 0.30 | ||
| Cluster 5 | 23.92 ± 2.39 | 2.42 ± 0.33 | ||
| Cluster 6 | 23.95 ± 2.59 | 2.37 ± 0.30 | ||
| Cluster 7 | 25.98 ± 5.27 | 2.43 ± 0.14 | ||
| Cluster 8 | 24.33 ± 2.64 | 2.35 ± 0.31 |
aData are presented as mean ± standard deviation
bNormality was assessed using the Shapiro–Wilk test; consequently, the Kruskal–Wallis test was applied to compare differences among groups
cThe length of the first mitosis = the time of the division resulting in 2 cells minus the time of PN disappearance
Table 4.
The relationship between each cluster and morphokinetic parameters until the 2-cell embryonic stage in the test dataset
| The onset of the first cleavage (hours)a | P-valueb | The length of the first mitosis (hours)a,c | P-valueb | |
|---|---|---|---|---|
| Cluster 0 | 24.45 ± 3.15 | 0.595 | 2.49 ± 0.51 | 0.937 |
| Cluster 1 | 23.91 ± 2.56 | 2.49 ± 0.45 | ||
| Cluster 2 | 23.30 ± 2.46 | 2.36 ± 0.25 | ||
| Cluster 3 | 25.17 ± 2.63 | 2.39 ± 0.26 | ||
| Cluster 4 | 23.45 ± 2.28 | 2.41 ± 0.49 | ||
| Cluster 5 | 23.76 ± 2.36 | 2.41 ± 0.38 | ||
| Cluster 6 | 24.15 ± 1.99 | 2.35 ± 0.30 | ||
| Cluster 7 | 23.44 ± 3.89 | 2.39 ± 0.13 | ||
| Cluster 8 | 23.30 ± 1.79 | 2.57 ± 0.68 |
aData are presented as mean ± standard deviation
bNormality was assessed using the Shapiro–Wilk test; consequently, the Kruskal–Wallis test was applied to compare differences among groups
cThe length of the first mitosis = the time of the division resulting in 2 cells minus the time of PN disappearance
The centroid trajectories of each cluster exhibited characteristic movement patterns (Supplemental Movies 1–9). Clusters 1, 3, and 8 were characterized by consistently low migration speeds with minimal fluctuations. Clusters 2, 4, 5, and 6 exhibited periodic increases in migration speed, with noticeable peaks occurring in frames 20–40 (3 to 5 h before PN disappearance). Cluster 7 showed a delayed but sharp increase in migration speed after frame 50 (2 h before PN disappearance), while Cluster 0 exhibited a random migration pattern without a clear trend.
Next, we investigated whether the live-birth rate following ART differed between the clusters. In the training dataset, a significant difference was observed in the live-birth rate between clusters (P < 0.001, Fig. 4a). The live-birth rate was significantly higher in Cluster 7 (100.0%, P < 0.01) and significantly lower in Cluster 2 (36.7%, P < 0.05). However, in the test dataset, no significant differences in live-birth rates were observed among the clusters (P = 0.184, Fig. 4b). Moreover, no significant differences in the live-birth rate were found between the training and test datasets for each cluster.
Fig. 4.
The live-birth rate in each of the nine clusters in the training and test datasets. Training dataset: Cluster 0, N = 25; Cluster 1, N = 18; Cluster 2, N = 30; Cluster 3, N = 17; Cluster 4, N = 22; Cluster 5, N = 30; Cluster 6, N = 27; Cluster 7, N = 8; and Cluster 8, N = 33 (a). Test dataset: Cluster 0, N = 17; Cluster 1, N = 17; Cluster 2, N = 12; Cluster 3, N = 12; Cluster 4, N = 16; Cluster 5, N = 13; Cluster 6, N = 24; Cluster 7, N = 3; and Cluster 8, N = 27 (b). Fisher’s exact probability test and residual analysis. **P-value < 0.01, * < 0.05
Because live-birth rates were similar between the training and test datasets for four of the nine clusters, we explored whether grouping the identified clusters would increase their predictive value. Therefore, we further categorized the time-series NPB migration speed clusters identified in the training data based on the live-birth rates in each cluster in the training dataset: low, live-birth rate ≤ 50.0%; middle, live-birth rate 50.1–65.0%; and high, live-birth rate > 65.0%. Clusters 1 and 7 comprised the high group; Clusters 0, 3, and 4 comprised the middle group, and Clusters 2, 5, 6, and 8 comprised the low group. As expected, the live-birth rate of the high subgroup was significantly higher than that of the low subgroup in the training data (76.9% vs. 45.8%, P = 0.014, Fig. 5a). We then examined live-birth rates in the grouped time-series clusters in the test data. In the test data, the live-birth rate in the grouped time-series clusters designated as high was significantly higher than that in the grouped time-series clusters designated as low (80.0% vs. 47.4%, P = 0.034, Fig. 5b) and demonstrated a tendency to be higher than that seen in the grouped clusters in the middle category (80.0% vs. 51.1%, P = 0.065, Fig. 5b). Thus, the live-birth rates in the test-data clusters closely mirrored those in the training-data clusters after grouping (Fig. 5).
Fig. 5.
The live-birth rate of the grouped clusters in the training and test datasets. Training dataset: low group, N = 120; middle group, N = 64; and high group, N = 26 (a). Test dataset: low group, N = 76; middle group, N = 45; and high group, N = 20 (b). Holm’s test. *P-value < 0.05, low group vs. high group
Discussion
This study demonstrated that time-series clustering analysis of the migration speed of NPB in human zygotes may serve as a predictor of live birth following ART. Even though changes in migration speed may seem irregular, changes in NPB migration speed over time could be classified into nine distinct temporal patterns. The live-birth rates in the test data closely mirrored those in the training data for four of the nine clusters. However, while there were no significant differences in the live-birth rates between the training and test data, the test data did not reflect the trends observed in the training data for five of the nine clusters. This may have been due to poor clustering quality, as indicated by a low silhouette score. To address this issue, we grouped the time-series clusters into three distinct categories based on the live-birth rates in the training data. After this grouping, the live-birth rate in the test dataset was correlated with the time-series clusters. We propose that time-series clustering may serve as an indicator for predicting live births following ART when the clusters are further consolidated into these three subgroups.
In this study, the live-birth rate in the test data was ~ 80% for zygotes classified into the high group based on NPB-migration time-series cluster, whereas live-birth rates were ~ 50% for zygotes classified into the middle or low groups. Single euploid vitrification-thawed blastocyst transfer cycles have yielded live-birth rates of approximately 60% [25, 26]. Using the clustering method that we developed, zygotes classified into the high group were expected to have a live-birth rate equal to or higher than that of zygotes undergoing preimplantation genetic testing for aneuploidy (PGT-A). However, the live-birth rates in zygotes classified into the middle or low groups were still relatively high, ranging from 47.4% to 51.1%. This suggests that using this classification as a criterion for embryo transfer or discard decisions might carry a high false-negative rate, potentially leading to the discarding of embryos with live-birth potential. Therefore, at this time, we must consider that this indicator has not been refined sufficiently for use in embryo-discard decisions, especially because only 14.2% of zygotes in the test dataset were classified into the high group. Although NPB migration speed and the pattern of NPB migration speed changes are potentially biomarkers for embryo selection, further validation is needed to confirm clinical reliability and reproducibility. Accordingly, this analysis of time-series changes in NPB migration speed suggests that this method may be useful for prioritizing embryo transfer and deciding whether to perform PGT-A.
Because NPBs co-localize with chromatin, the position of NPBs can serve as a proxy for chromatin distribution in the zygote [7, 8]. Chromatin migration has been observed across species, including yeast [27], Drosophila [28], and humans [29–31], and is regulated by nucleosome dynamics [31]. In mammalian nuclei, chromatin is more mobile in early G1 but becomes relatively stable during mid-G1, late G1, S, and G2 phases [32]. Increased transcriptional activity can promote chromatin mobility and repositioning [32]. Chromatin dynamics are restricted by active RNA polymerase II (RNAPII) [33], while RNAPII inhibition increases mobility [34]. Despite ongoing transcription in mid- and late G1, chromatin movement remains limited [32]. Given that NPB movement indirectly reflects chromatin dynamics, we hypothesize that more dynamic NPB movement may indicate the conclusion of an efficient transcriptional process, thereby supporting proper chromatin organization and promoting euploidy and live birth. Chromosomal aneuploidy in human embryos is typically acquired during mitosis [35]. Nucleolar protein complexes are essential for mitotic chromosome segregation and ensure their stability [36]. Nucleoplasmin 2, a key nucleolar protein, is required for the reconstitution of NPBs and plays a critical role in organizing centromeric and pericentromeric chromatin [37]. Furthermore, the oocyte nucleolus structure formed by nucleoplasmin 2 provides a foundation for correct pericentromeric heterochromatin organization, which is crucial for accurate chromosome segregation [37]. Thus, structural abnormalities in NPBs may lead to defective chromosome segregation, resulting in aneuploidy.
Accordingly, we propose that the speed and patterns of NPB migration reflect the structural integrity of NPBs. In this study, time-series clustering of NPB migration speed in human zygotes was found to be a potential predictor of live birth following ART. Stable migration speed patterns (as seen in Cluster 1) and late-onset acceleration (as seen in Cluster 7) may be associated with higher live-birth rates, while excessive periodic fluctuations (seen in Cluster 2) or irregular movements (seen in Cluster 5) may be linked with lower live-birth rates. Previous research showed that NPBs migrated more slowly in aneuploid embryos and in embryos from patients who experienced miscarriage or implantation failure, as compared with those resulting in live birth. These findings, when coupled with previous works [22, 23], suggest that NPB migration speed and the pattern may serve as an indicator of NPB structural normality and, by extension, reflect the ploidy status of the embryo.
We previously demonstrated that zygotes with faster NPB migration speed possess the developmental potential for live birth, and the migration speed of mNPBs is a novel predictor of ploidy status and live birth following ART [22, 23]. However, that analysis of the NPB migration speed had several disadvantages. Kinovea motion capture software could not completely automatically track NPBs, and their central coordinates have to be manually confirmed and revised in every image. We attempted to develop tracking software, but the software was unable to automatically track NPBs efficiently when focusing on them with time-lapse microscopy because the image quality was grainy. Additionally, during NPB tracking, both of the software systems we tested often tracked an NPB that was different from the intended target or mistakenly identified cytoplasmic granules as NPBs. As a result, the analysis of a single NPB took approximately 10–15 min (mean analysis frames: 117.4 ± 42.3 per NPB), and the analysis of a single zygote took approximately 60–90 min (three NPBs were analyzed per PN). Therefore, this analysis method was not practical for clinical application. In the current analysis, we focused on the period from 1 to 7 h (60 frames) before PN disappearance, and the number of NPBs analyzed was limited to one per zygote, theoretically reducing the analysis time to 5.0–7.5 min per zygote. Furthermore, unlike commercially available software for predicting embryo development, our analysis code was custom-made, so the accuracy of the analysis will likely improve as more data is accumulated. Advances in image processing, including artificial intelligence-assisted tracking systems and high-resolution imaging, hold great promise for significantly streamlining the analysis process in future studies.
Our results showed that the migration speed of mNPBs was significantly faster in ICSI-derived zygotes than in IVF-derived ones. This may reflect fundamental differences between the two fertilization processes. During ICSI, the acrosome (which contains hydrolytic enzymes) and the sperm plasma membrane are introduced directly into the ooplasm [38]. Additionally, oocyte activation patterns, as indicated by calcium oscillations, differ between IVF and ICSI [39]. Moreover, DNA replication in the male PN—but not in the female PN—is known to be delayed following ICSI [40]. These factors may collectively alter the cytoplasmic environment surrounding the male PN, thereby influencing nuclear dynamics, including NPB migration.
Furthermore, our data demonstrated that, in both ICSI- and IVF-derived zygotes, the migration speed of NPBs in male PNs was significantly faster than that in female PNs. In ICSI-derived zygotes, the male PN contained a larger number of NPBs than the female PN [21], and a similar trend was observed in IVF-derived zygotes. In this study, we analyzed only representative NPBs, selected based on consistent motility and prolonged visibility. Upon re-examination of the zygotes used in the analysis, we found that the number of analyzable mNPBs was greater than that of fNPBs in most zygotes. Specifically, the mean numbers of mNPBs and fNPBs per PN were 7.6 ± 2.2 and 4.9 ± 2.1, respectively. In contrast, 0.6 ± 1.7 NPBs per zygote could not be analyzed due to overlapping PNs (unpublished data). Such differences in NPB number may partly explain the consistently higher migration speed observed in the male PN. This potential selection bias is particularly relevant because NPBs were selected for analysis based on their prolonged appearance and consistently high motility throughout the observation period.
Previous studies have reported that sperm chromatin remodeling after ICSI proceeds more asynchronously than after IVF, leading to differences in chromatin decondensation and the timing of DNA synthesis in mice [41]. Additionally, chromatin dynamics differ between the male and female PNs in the mouse zygote as a result of parental competition: the chromatin of the female PN is significantly less dynamic than that of the male PN derived from the sperm [42]. Since NPBs are known to colocalize with chromatin [7, 8], the observed disparities in NPB migration speed may be attributable to two factors: differences in chromatin dynamics between male and female PNs and fertilization-specific chromatin remodeling.
This study has some limitations. First, the sample size is relatively small, particularly in the test dataset, which may have limited our ability to detect statistically significant differences in live-birth rates among the clusters. Second, four of the nine clusters were classified into the low group, which accounted for 57.1% of all zygotes in the test dataset. If the low group proves to be a good indicator for predicting no-pregnancy rates, there may be an opportunity to create a more refined predictive model for live birth following ART to create human zygotes by further subdividing this group and analyzing its characteristics. Third, the clustering method may need refinement to account for other variables not included in this analysis, such as maternal age, embryo quality, and other clinical factors. Future research should focus on expanding the dataset and incorporating additional predictive markers to better understand the relationship between NPB dynamics and embryo viability. Additionally, we plan to further investigate the relationship between time-series clustering of NPB migration patterns and differences in fertilization methods. By refining our clustering models, we may be able to develop a more accurate and clinically relevant tool for improving clinical outcomes following ART. Collaborative efforts across multiple centers and the integration of additional clinical factors, such as maternal age and embryo morphology, could enhance the predictive accuracy of clustering models.
In conclusion, the patterns of NPB migration speed over time could serve as a useful marker for selecting embryos with a higher potential for developing into viable pregnancies and live births following ART. Time-series analysis of migration using clustering methodology is a promising approach to quantify differences in NPB migration speed as embryonic development progresses prior to embryo implantation.
Supplementary Information
Below is the link to the electronic supplementary material.
Typical tracking of a nucleolus precursor body (NPB) from 1 to 7 h before pronuclear (PN) disappearance in Cluster 0. Please note that this movie shows NPB tracking without correction with PN coordinates, and thus does not fully reflect the migration speed pattern used in the clustering analysis (MP4 9.72 MB)
Typical tracking of a nucleolus precursor body (NPB) from 1 to 7 h before pronuclear (PN) disappearance in Cluster 1. Please note that this movie shows NPB tracking without correction with PN coordinates, and thus does not fully reflect the migration speed pattern used in the clustering analysis (MP4 9.55 MB)
Typical tracking of a nucleolus precursor body (NPB) from 1 to 7 h before pronuclear (PN) disappearance in Cluster 2. Please note that this movie shows NPB tracking without correction with PN coordinates, and thus does not fully reflect the migration speed pattern used in the clustering analysis (MP4 9.65 MB)
Typical tracking of a nucleolus precursor body (NPB) from 1 to 7 h before pronuclear (PN) disappearance in Cluster 3. Please note that this movie shows NPB tracking without correction with PN coordinates, and thus does not fully reflect the migration speed pattern used in the clustering analysis (MP4 9.78 MB)
Typical tracking of a nucleolus precursor body (NPB) from 1 to 7 h before pronuclear (PN) disappearance in Cluster 4. Please note that this movie shows NPB tracking without correction with PN coordinates, and thus does not fully reflect the migration speed pattern used in the clustering analysis (MP4 9.79 MB)
Typical tracking of a nucleolus precursor body (NPB) from 1 to 7 h before pronuclear (PN) disappearance in Cluster 5. Please note that this movie shows NPB tracking without correction with PN coordinates, and thus does not fully reflect the migration speed pattern used in the clustering analysis (MP4 10.3 MB)
Typical tracking of a nucleolus precursor body (NPB) from 1 to 7 h before pronuclear (PN) disappearance in Cluster 6. Please note that this movie shows NPB tracking without correction with PN coordinates, and thus does not fully reflect the migration speed pattern used in the clustering analysis (MP4 9.78 MB)
Typical tracking of a nucleolus precursor body (NPB) from 1 to 7 h before pronuclear (PN) disappearance in Cluster 7. Please note that this movie shows NPB tracking without correction with PN coordinates, and thus does not fully reflect the migration speed pattern used in the clustering analysis (MP4 10.0 MB)
Typical tracking of a nucleolus precursor body (NPB) from 1 to 7 h before pronuclear (PN) disappearance in Cluster 8. Please note that this movie shows NPB tracking without correction with PN coordinates, and thus does not fully reflect the migration speed pattern used in the clustering analysis (MP4 9.63 MB)
Acknowledgements
We thank Shogo Shiratsuki PhD (Merck Biopharma Co., Ltd.) for the critical review and scientific discussion of the manuscript. We thank Shannon Wyszomierski PhD for editing the manuscript.
Author contribution
T. I. designed the study, developed the programming code, analyzed the data, and drafted the manuscript. T. I. and S. T. annotated morphokinetic parameters. M. U. critically reviewed the manuscript. Y. Y. obtained informed consent. All authors have approved the final manuscript.
Declarations
Ethical approval
All procedures followed were in accordance with the ethical standards set forth by the appropriate committee responsible for human experimentation (institutional and national) and with the Helsinki Declaration of 1964 and its later amendments. Informed consent was obtained from all patients to be included in the study. This study was approved by the Umeda Fertility Clinic Institutional Review Board (181215R2).
Conflict of interest
The authors declare no competing interests.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Berry J, Weber SC, Vaidya N, Haataja M, Brangwynne CP. RNA transcription modulates phase transition-driven nuclear body assembly. Proc Natl Acad Sci U S A. 2015;112:E5237–45. 10.1073/pnas.1509317112. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Kyogoku H, Kitajima TS, Miyano T. Nucleolus precursor body (NPB): a distinct structure in mammalian oocytes and zygotes. Nucleus. 2014;5:493–8. 10.4161/19491034.2014.990858. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Fulka H, Rychtarova J, Loi P. The nucleolus-like and precursor bodies of mammalian oocytes and embryos and their possible role in post-fertilization centromere remodelling. Biochem Soc Trans. 2020;48:581–93. 10.1042/BST20190847. [DOI] [PubMed] [Google Scholar]
- 4.Fulka H, Aoki F. Nucleolus precursor bodies and ribosome biogenesis in early mammalian embryos: old theories and new discoveries. Biol Reprod. 2016;94:143. 10.1095/biolreprod.115.136093. [DOI] [PubMed] [Google Scholar]
- 5.Kresoja-Rakic J, Santoro R. Nucleolus and rRNA gene chromatin in early embryo development. Trends Genet. 2019;35:868–79. 10.1016/j.tig.2019.06.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Ogushi S, Palmieri C, Fulka H, Saitou M, Miyano T, Fulka J Jr. The maternal nucleolus is essential for early embryonic development in mammals. Science. 2008;319:613–6. 10.1126/science.1151276. [DOI] [PubMed] [Google Scholar]
- 7.Coticchio G, Borini A, Albertini DF. The slippery slope antedating syngamy: pronuclear activity in preparation for the first cleavage. J Assist Reprod Genet. 2021;38:1721–3. 10.1007/s10815-021-02238-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Cavazza T, Takeda Y, Politi AZ, Aushev M, Aldag P, Baker C, et al. Parental genome unification is highly error-prone in mammalian embryos. Cell. 2021;184:2860–77. 10.1016/j.cell.2021.04.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Otsuki J, Iwasaki T, Tsuji Y, Katada Y, Sato H, Tsutsumi Y, et al. Potential of zygotes to produce live births can be identified by the size of the male and female pronuclei just before their membranes break down. Reprod Med Biol. 2017;16(2):200–5. 10.1002/rmb2.12032. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Otsuki J, Iwasaki T, Enatsu N, Katada Y, Furuhashi K, Shiotani M. Noninvasive embryo selection: kinetic analysis of female and male pronuclear development to predict embryo quality and potential to produce live birth. Fertil Steril. 2019;112(5):874–81. 10.1016/j.fertnstert.2019.07.015. [DOI] [PubMed] [Google Scholar]
- 11.Araki E, Itoi F, Honnma H, Asano Y, Oguri H, Nishikawa K. Correlation between the pronucleus size and the potential for human single pronucleus zygotes to develop into blastocysts: 1PN zygotes with large pronuclei can expect an embryo development to the blastocyst stage that is similar to the development of 2PN zygotes. J Assist Reprod Genet. 2018;35(5):817–23. 10.1007/s10815-018-1137-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Coticchio G, Mignini Renzini M, Novara PV, Lain M, De Ponti E, Turchi D, et al. Focused time-lapse analysis reveals novel aspects of human fertilization and suggests new parameters of embryo viability. Hum Reprod. 2018;33(1):23–31. 10.1093/humrep/dex344. [DOI] [PubMed] [Google Scholar]
- 13.Porokh V, Kyjovská D, Martonová M, Klenková T, Otevřel P, Kloudová S, et al. Zygotic spindle orientation defines cleavage pattern and nuclear status of human embryos. Nat Commun. 2024;15(1): 6369. 10.1038/s41467-024-50732-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Tesarik J, Greco E. The probability of abnormal preimplantation development can be predicted by a single static observation on pronuclear stage morphology. Hum Reprod. 1999;14(5):1318–23. 10.1093/humrep/14.5.1318. [DOI] [PubMed] [Google Scholar]
- 15.Scott L. Pronuclear scoring as a predictor of embryo development. Reprod Biomed Online. 2003;6(2):201–14. 10.1016/s1472-6483(10)61711-7. [DOI] [PubMed] [Google Scholar]
- 16.Depa-Martynow M, Jedrzejczak P, Pawelczyk L. Pronuclear scoring as a predictor of embryo quality in in vitro fertilization program. Folia Histochem Cytobiol. 2007;45(Suppl 1):S85–9. [PubMed] [Google Scholar]
- 17.Balaban B, Yakin K, Urman B, Isiklar A, Tesarik J. Pronuclear morphology predicts embryo development and chromosome constitution. Reprod Biomed Online. 2004;8(6):695–700. 10.1016/s1472-6483(10)61651-3. [DOI] [PubMed] [Google Scholar]
- 18.Fulka H, Kyogoku H, Zatsepina O, Langerova A, Fulka J Jr. Can nucleoli be markers of developmental potential in human zygotes? Trends Mol Med. 2015;21(11):663–72. 10.1016/j.molmed.2015.09.005. [DOI] [PubMed] [Google Scholar]
- 19.Berger DS, Zapantis A, Merhi Z, Younger J, Polotsky AJ, Jindal SK. Embryo quality but not pronuclear score is associated with clinical pregnancy following IVF. J Assist Reprod Genet. 2014;31(3):279–83. 10.1007/s10815-013-0162-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Mio Y. Morphological analysis of human embryonic development using time-lapse cinematography. J Mamm Ova Res. 2006;23:27–35. 10.1274/jmor.23.27. [Google Scholar]
- 21.Montag M, Liebenthron J, Köster M. Which morphological scoring system is relevant in human embryo development? Placenta. 2011;32(Suppl 3):S252–6. 10.1016/j.placenta.2011.07.009. [DOI] [PubMed] [Google Scholar]
- 22.Inoue T, Taguchi S, Uemura M, Tsujimoto Y, Miyazaki K, Yamashita Y. Migration speed of nucleolus precursor bodies in human male pronuclei: a novel parameter for predicting live birth. J Assist Reprod Genet. 2021;38:1725–36. 10.1007/s10815-021-02172-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Inoue T, Taguchi S, Uemura M, Tsujimoto Y, Kokunai K, Ikawa K, et al. The migration speed of nucleolar precursor bodies in pronuclei affects in vitro fertilization-derived human embryo ploidy status and live birth. Reprod Med Biol. 2023;22(1): e12497. 10.1002/rmb2.12497. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Kanda Y. Investigation of the freely available easy-to-use software ‘EZR’ for medical statistics. Bone Marrow Transplant. 2013;48:452–8. 10.1038/bmt.2012.244. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Zhang L, Wei D, Zhu Y, Gao Y, Yan J, Chen ZJ. Rates of live birth after mosaic embryo transfer compared with euploid embryo transfer. J Assist Reprod Genet. 2019;36(1):165–72. 10.1007/s10815-018-1322-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Boynukalin FK, Gultomruk M, Cavkaytar S, Turgut E, Findikli N, Serdarogullari M, et al. Parameters impacting the live birth rate per transfer after frozen single euploid blastocyst transfer. PLoS One. 2020;15(1): e0227619. 10.1371/journal.pone.0227619. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Cheblal A, Challa K, Seeber A, Shimada K, Yoshida H, Ferreira HC, et al. DNA damage-induced nucleosome depletion enhances homology search independently of local break movement. Mol Cell. 2020;80:311-26.e4. 10.1016/j.molcel.2020.09.002. [DOI] [PubMed] [Google Scholar]
- 28.Vazquez J, Belmont AS, Sedat JW. Multiple regimes of constrained chromosome motion are regulated in the interphase Drosophila nucleus. Curr Biol. 2001;11:1227–39. 10.1016/S0960-9822(01)00390-6. [DOI] [PubMed] [Google Scholar]
- 29.Chubb JR, Boyle S, Perry P, Bickmore WA. Chromatin motion is constrained by association with nuclear compartments in human cells. Curr Biol. 2002;12:439–45. 10.1016/S0960-9822(02)00695-4. [DOI] [PubMed] [Google Scholar]
- 30.Walter J, Schermelleh L, Cremer M, Tashiro S, Cremer T. Chromosome order in HeLa cells changes during mitosis and early G1, but is stably maintained during subsequent interphase stages. J Cell Biol. 2003;160:685–97. 10.1083/jcb.200211103. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Ashwin SS, Nozaki T, Maeshima K, Sasai M. Organization of fast and slow chromatin revealed by single-nucleosome dynamics. Proc Natl Acad Sci U S A. 2019;116:19939–44. 10.1073/pnas.1907342116. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Dion V, Gasser SM. Chromatin movement in the maintenance of genome stability. Cell. 2013;152:1355–64. 10.1016/j.cell.2013.02.010. [DOI] [PubMed] [Google Scholar]
- 33.Hochberg-Laufer H, Shav-Tal Y. Active RNA polymerase II curbs chromatin movement. J Cell Biol. 2019;218:1427–8. 10.1083/jcb.201903050. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Nagashima R, Hibino K, Ashwin SS, Babokhov M, Fujishiro S, Imai R, et al. Single nucleosome imaging reveals loose genome chromatin networks via active RNA polymerase II. J Cell Biol. 2019;218:1511–30. 10.1083/jcb.201811090. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Palmerola KL, Amrane S, De Los AA, Xu S, Wang N, de Pinho J, et al. Replication stress impairs chromosome segregation and preimplantation development in human embryos. Cell. 2022;185:2988-3007.e20. 10.1016/j.cell.2022.06.028. [DOI] [PubMed] [Google Scholar]
- 36.Fujimura A, Hayashi Y, Kato K, Kogure Y, Kameyama M, Shimamoto H, et al. Identification of a novel nucleolar protein complex required for mitotic chromosome segregation through centromeric accumulation of Aurora B. Nucleic Acids Res. 2020;48:6583–96. 10.1093/nar/gkaa449. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Ogushi S, Yamagata K, Obuse C, Furuta K, Wakayama T, Matzuk MM, et al. Reconstitution of the oocyte nucleolus in mice through a single nucleolar protein, NPM2. J Cell Sci. 2017;130:2416–29. 10.1242/jcs.195875. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Morozumi K, Shikano T, Miyazaki S, Yanagimachi R. Simultaneous removal of sperm plasma membrane and acrosome before intracytoplasmic sperm injection improves oocyte activation/embryonic development. Proc Natl Acad Sci U S A. 2006;103(47):17661–6. 10.1073/pnas.0608183103. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Kurokawa M, Fissore RA. ICSI-generated mouse zygotes exhibit altered calcium oscillations, inositol 1,4,5-trisphosphate receptor-1 down-regulation, and embryo development. Mol Hum Reprod. 2003;9(9):523–33. 10.1093/molehr/gag072. [DOI] [PubMed] [Google Scholar]
- 40.Terada Y, Luetjens CM, Sutovsky P, Schatten G. Atypical decondensation of the sperm nucleus, delayed replication of the male genome, and sex chromosome positioning following intracytoplasmic human sperm injection (ICSI) into golden hamster eggs: does ICSI itself introduce chromosomal anomalies? Fertil Steril. 2000;74(3):454–60. 10.1016/s0015-0282(00)00671-3. [DOI] [PubMed] [Google Scholar]
- 41.Ajduk A, Yamauchi Y, Ward MA. Sperm chromatin remodeling after intracytoplasmic sperm injection differs from that of in vitro fertilization. Biol Reprod. 2006;75(3):442–51. 10.1095/biolreprod.106.053223. [DOI] [PubMed] [Google Scholar]
- 42.Ooga M, Inoue R, Kazama K, Wakayama S, Kamimura S, Wakayama T. Parental competition for the regulators of chromatin dynamics in mouse zygotes. Commun Biol. 2022;5(1):699. 10.1038/s42003-022-03623-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Typical tracking of a nucleolus precursor body (NPB) from 1 to 7 h before pronuclear (PN) disappearance in Cluster 0. Please note that this movie shows NPB tracking without correction with PN coordinates, and thus does not fully reflect the migration speed pattern used in the clustering analysis (MP4 9.72 MB)
Typical tracking of a nucleolus precursor body (NPB) from 1 to 7 h before pronuclear (PN) disappearance in Cluster 1. Please note that this movie shows NPB tracking without correction with PN coordinates, and thus does not fully reflect the migration speed pattern used in the clustering analysis (MP4 9.55 MB)
Typical tracking of a nucleolus precursor body (NPB) from 1 to 7 h before pronuclear (PN) disappearance in Cluster 2. Please note that this movie shows NPB tracking without correction with PN coordinates, and thus does not fully reflect the migration speed pattern used in the clustering analysis (MP4 9.65 MB)
Typical tracking of a nucleolus precursor body (NPB) from 1 to 7 h before pronuclear (PN) disappearance in Cluster 3. Please note that this movie shows NPB tracking without correction with PN coordinates, and thus does not fully reflect the migration speed pattern used in the clustering analysis (MP4 9.78 MB)
Typical tracking of a nucleolus precursor body (NPB) from 1 to 7 h before pronuclear (PN) disappearance in Cluster 4. Please note that this movie shows NPB tracking without correction with PN coordinates, and thus does not fully reflect the migration speed pattern used in the clustering analysis (MP4 9.79 MB)
Typical tracking of a nucleolus precursor body (NPB) from 1 to 7 h before pronuclear (PN) disappearance in Cluster 5. Please note that this movie shows NPB tracking without correction with PN coordinates, and thus does not fully reflect the migration speed pattern used in the clustering analysis (MP4 10.3 MB)
Typical tracking of a nucleolus precursor body (NPB) from 1 to 7 h before pronuclear (PN) disappearance in Cluster 6. Please note that this movie shows NPB tracking without correction with PN coordinates, and thus does not fully reflect the migration speed pattern used in the clustering analysis (MP4 9.78 MB)
Typical tracking of a nucleolus precursor body (NPB) from 1 to 7 h before pronuclear (PN) disappearance in Cluster 7. Please note that this movie shows NPB tracking without correction with PN coordinates, and thus does not fully reflect the migration speed pattern used in the clustering analysis (MP4 10.0 MB)
Typical tracking of a nucleolus precursor body (NPB) from 1 to 7 h before pronuclear (PN) disappearance in Cluster 8. Please note that this movie shows NPB tracking without correction with PN coordinates, and thus does not fully reflect the migration speed pattern used in the clustering analysis (MP4 9.63 MB)