iiinova — where your digital self keeps evolving

Matt Ming @mattjming.bsky.social · 07/01/2026

I want to thank my advisors and other members of the @arbelharpak.bsky.social lab for their helpful feedback and support! (12/12)

000

Matt Ming @mattjming.bsky.social · 07/01/2026

Finally, although our methodology attempts to address study-specific participation bias, systemic sex-specific participation biases may still persist across studies and lead to observed allele frequency differences. (11/12)

120

Matt Ming @mattjming.bsky.social · 07/01/2026

Viability selection occurring over one generation has canonically been hypothesized to drive allele frequency divergence. We find evidence for several genes associated with diseases impacting viability during development and adulthood. (10/12)

100

Matt Ming @mattjming.bsky.social · 07/01/2026

One cool example: genes expressed in sperm and affecting spermatogenesis. We hypothesize that allelic effects on fertilization (e.g., via cell mobility, clonal expansion doi.org/10.1038/s415...) may differ between X and Y sperm (9/12)

110

Matt Ming @mattjming.bsky.social · 07/01/2026

Through additional analysis, including within-sex deviations from Hardy-Weinberg Equilibrium (that can provide evidence for selection private to one sex vs. acting in both sexes), we ask which of the hypotheses are most plausible for each cross-study candidates. (8/12)

120

Matt Ming @mattjming.bsky.social · 07/01/2026

We investigate hypotheses for drivers of male-female allele frequency differences, including four hypotheses: 1) differential selection on X- and Y-carrying sperm, 2-3) viability selection during fetal development and post-birth, and 4) through genetic study participation bias. (7/12)

110

Matt Ming @mattjming.bsky.social · 07/01/2026

Intriguingly, ADRA1B has been implicated in cancer and sexually dimorphic response to insulin in mice. ANKRD36/ANKDR36C are primarily expressed in sperm and appear to play a role in male fertility. (6/12)

100

Matt Ming @mattjming.bsky.social · 07/01/2026

Further analysis of the robustness of evidence across studies identifies 3 genes—ANKRD36, ANKRD36C, and ADRA1B—as being the strongest candidate targets of sex-differential selection. (5/12)

110

Matt Ming @mattjming.bsky.social · 07/01/2026

We identify 12 genes with significant between-sex allele frequency divergence which replicates across studies (4/12)

110

Matt Ming @mattjming.bsky.social · 07/01/2026

Previous work, including by us, has focused on single genetic studies such as the UK Biobank. Here, we analyze the replication of allele frequency divergences from three biobank-scale studies—gnomAD, UKB, and All of Us (3/12) doi.org/10.1073/pnas... doi.org/10.1371/jour... doi.org/10.1016/j.xg...

110

Matt Ming @mattjming.bsky.social · 07/01/2026

Sex differences in allele frequencies may be driven by sex-differential selection. However, this signal may be confounded by participation biases and bioinformatic artifacts, some of which are study specific. (2/12)

100

Matt Ming @mattjming.bsky.social · 07/01/2026

I’m very excited to share our new preprint with @arbelharpak.bsky.social! We meta-analyze male-female allele frequency divergences across studies (gnomAD, UKB, AoU) and ask what drives the observed differences. (1/12) www.biorxiv.org/content/10.6...

185

Matt Ming @mattjming.bsky.social · 14/01/2025

I want to shout out @aneil-agrawal.bsky.social, @jmillercole.bsky.social and the @arbelharpak.bsky.social lab for helpful feedback and discussion! I also sincerely thank our AJHG editors and reviewers who gave me an overwhelmingly positive (and dare I say fun) first publication experience! (14/14)

030

Matt Ming @mattjming.bsky.social · 14/01/2025

This project has been an incredibly rewarding and educational experience for me, and I’m thankful to my coauthor Changde Cheng and my advisors Mark Kirkpatrick and @arbelharpak.bsky.social for their work, thoughts, and guidance throughout! (13/14)

110

Matt Ming @mattjming.bsky.social · 14/01/2025

We conclude that while causal relationships between SDS and sex-differential gene expression in humans remain plausible, they are yet to be fully elucidated. We are excited to continue exploring how signals and drivers of genome-wide SDS may be detected. (12/14)

100

Matt Ming @mattjming.bsky.social · 14/01/2025

We also discuss how other factors may lead to a lack of signal. For example, sex-differential expression driving SDS may be rare. Or sex-differential expression may be related to past but not current SDS. (11/14)

100

Matt Ming @mattjming.bsky.social · 14/01/2025

We hypothesize eQTL ascertainment bias, as explored by @hakha.bsky.social et al. could contribute to the lack of signal: eQTL discovery is biased against sites under strong selection, weakening a relationship between FST and sex-differential expression (10/14) doi.org/10.1038/s415...

100

Matt Ming @mattjming.bsky.social · 14/01/2025

However, work by Carrie Zhu, @jmillercole.bsky.social , @filipruzicka.bsky.social, and others have shown pervasive genome-wide signals of SDS using other methods. (10/14) doi.org/10.1016/j.xg... doi.org/10.1073/pnas... doi.org/10.1371/jour...

100

Matt Ming @mattjming.bsky.social · 14/01/2025

Our results indicate that, even if a relationship between SDS and sex-differential expression exists, its signals are difficult to model and detect. (9/14)

100

Matt Ming @mattjming.bsky.social · 14/01/2025

Alas, across 34 human tissues, we find no evidence for a general, genome-wide relationship between sex-differential expression and SDS. (8/14)

100

Matt Ming @mattjming.bsky.social · 14/01/2025

We next re-investigate the relationship between between-sex FST and sex-differential expression with new and improved data (thanks to 2025 > 2016) and statistical model. (7/14)

100

Matt Ming @mattjming.bsky.social · 14/01/2025

We thus revisit the interpretation that a Twin Peaks pattern suggests sex-differential gene expression drives SDS. We find 21.2% of empirical null samples result in Twin Peaks, so Twin Peaks is consistent with no relationship between sex-differential expression and SDS. (6/14)

100

Matt Ming @mattjming.bsky.social · 14/01/2025

In our manuscript, we found caveats to the Twin Peaks theoretical expectation. (5/14)

100

Matt Ming @mattjming.bsky.social · 14/01/2025

Cheng and Kirkpatrick explicitly modeled the relationship between SDS and sex-differential expression and describe a “Twin Peaks” pattern which arises when sex-differential expression drives SDS. (4/14) doi.org/10.1371/jour...

100

Matt Ming @mattjming.bsky.social · 14/01/2025

Previous work has observed a relationship between SDS and sex differences in gene expression. However, these results have been hotly contested. (3/14)

100

Matt Ming @mattjming.bsky.social · 14/01/2025

Sex-differential selection (SDS) occurs when allelic fitness effects differ between males and females. SDS acting on viability will lead to differences in allele frequencies between the sexes, quantified using between-sex FST. (2/14)

110

Matt Ming @mattjming.bsky.social · 14/01/2025

I am excited to share the first first-author paper of my PhD describing work with Changde Cheng, Mark Kirkpatrick and @arbelharpak.bsky.social has been published at AJHG! We ask if sex-differential gene expression drives sex-differential selection in humans. (1/14) www.cell.com/ajhg/fulltex...

cell.com

No evidence for sex-differential transcriptomes driving genome-wide sex-differential natural selection

We assess the evidence for a genome-wide relationship between sex differences in gene expression and sex differences in natural selection. We develop an improved model for testing this association but...

186