This week’s recap highlights analysis of human de novo mutation rates from a four-generation pedigree reference, how LLMs internalize scientific literature and citation practices, the py_ped_sim forward pedigree and genetic simulator for complex family pedigree analysis, and a review on predicting gene expression from DNA sequence using deep learning models like Enformer and Borzoi.

In the previous post, I introduced the N=N double bond in nitrosobenzene dimer, arguing that even though it was a formal double bond, its bond dissociation energy made it nonetheless a very weak double bond! This was backed up by a technique known as energy decomposition analysis or EDA.

Do you recognize this? You read something and then you think something else because of that, and this Something is so banal, so self-evident, so obvious – and then you want to quote the relevant standard work (book, article, ISO standard), but you can’t find anything? Has this self-evident fact really never been written down before?

Kennt Ihr das? Ihr lest etwas, und denkt etwas anderes dazu, und dieses etwas ist so banal, so selbstverständlich, so offensichtlich – und dann wollt Ihr dazu das relevante Standardwerk (Buch, Aufsatz, ISO-Norm) zitieren, aber ihr findet nichts? Wurde diese Selbstverständlichkeit wirklich bisher noch von niemandem aufgeschrieben?

My old blog had (has) comments via the Blogger.com platform, but I did not have anything for the new blog. A couple of options are used, like Disqus and comments via GitHub. However, these both have the downside of a vendor lock-in and the whole point of moving my blog was to break out from such lock-ins.

Last week, I was interviewed on the program Breitband on Deutschlandfunk Kultur about the ruling of a Belgian court against four shadow libraries, as well as the Open Library of the Internet Archive (Krempl, 2025). The broadcast is now available online. Enjoy listening! Further information about the research group can be found on our official website.

At the frontier of technology, humanity is the experiment. The question now is, who is designing the experiment?

Leiden Madtrics readers might already be familiar with what some have called a “great project” but a “terrible acronym”: GLOBAL, the Guidance List for the repOrting of Bibliometric AnaLyses. Last summer, we invited bibliometricians to join the GLOBAL Delphi study to co-develop a reporting guideline for bibliometric analyses.

The Rogue Scholar science blog archive has introduced authentication with passkeys and will disable local accounts on September 15. Reading Rogue Scholar content has always been free and never required user accounts or cookie permissions. This is true both for web and API usage.

Für Bibliotheken und Archive gelten jeweils unterschiedliche rechtliche Anforderungen. Was gilt aber für Einrichtungen, die Archiv und Bibliothek zugleich sind? Der neue von Paul Klimpel und Katrin Lehnert verfasste Leitfaden des Digitalen Deutschen Frauenarchivs befasst sich mit dieser Frage und gibt Antworten. Kleinere Institutionen beheimaten häufig Orte für Recherche, die beides sind: Bibliothek und Archiv.

This post is based on a presentation I gave in June 2025 as part of a Metascience 2025 Preconference Virtual Symposium convened by Sven Ulpts and Sheena Bartscherer and including Thomas Hostler, Lai Ma, Lisa Malich, and Carlos Santana.