<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Psychiatric-Genomics on BIOCOMPUTER</title>
    <link>https://biocomputer.com/tags/psychiatric-genomics/</link>
    <description>Recent content in Psychiatric-Genomics on BIOCOMPUTER</description>
    <generator>Hugo</generator>
    <language>en-us</language>
    <lastBuildDate>Wed, 08 Apr 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://biocomputer.com/tags/psychiatric-genomics/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>1.14 Billion Rows of Psychiatric DNA — Now One Line of Python Away</title>
      <link>https://biocomputer.com/blog/pgc-gwas-psychiatric-genomics-huggingface/</link>
      <pubDate>Wed, 08 Apr 2026 00:00:00 +0000</pubDate>
      <guid>https://biocomputer.com/blog/pgc-gwas-psychiatric-genomics-huggingface/</guid>
      <description>&lt;p&gt;On April 7, 2026, Maziyar Panahi of &lt;strong&gt;OpenMed_AI&lt;/strong&gt; posted what might be the most consequential single data release in computational psychiatry this year. Every &lt;strong&gt;GWAS summary statistic&lt;/strong&gt; ever published by the &lt;strong&gt;Psychiatric Genomics Consortium&lt;/strong&gt; — 1.14 billion rows, 52 publications, 12 major disorder groups — is now sitting on Hugging Face in clean Apache Parquet format, queryable with a single line of Python.&lt;/p&gt;&#xA;&lt;p&gt;Before this, accessing PGC data meant chasing scattered Figshare links, wrangling inconsistent column separators, and running &lt;code&gt;wget | gunzip | awk&lt;/code&gt; pipelines for each of 52 datasets. Researchers weren&amp;rsquo;t doing science. They were doing plumbing.&lt;/p&gt;&#xA;&lt;p&gt;The barrier wasn&amp;rsquo;t intellectual. It was logistical. And OpenMed_AI just demolished it.&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
