End-to-end differentiable learning of protein structure

Mohammed AlQuraishi

doi:10.1101/265231

CARVIEW

MOTORHOMES

Select Language

HTTP/2 301 date: Sun, 18 Jan 2026 20:39:23 GMT content-length: 0 location: https://doi.org/10.1101/265231 server: cloudflare vary: Origin expires: Mon, 19 Jan 2026 20:39:22 GMT permissions-policy: interest-cohort=(),browsing-topics=() cf-cache-status: DYNAMIC nel: {"report_to":"cf-nel","success_fraction":0.0,"max_age":604800} strict-transport-security: max-age=31536000; includeSubDomains; preload report-to: {"group":"cf-nel","max_age":604800,"endpoints":[{"url":"https://a.nel.cloudflare.com/report/v4?s=cm2jil76ZGb5AyC8RnHHemlBYxi2U%2Bn00CjKUmGT43YYeQ5sc4t1fkHLjN8t68L9LvHzmovkzwnOtDZsSwxoe02z%2F%2BQr9A%3D%3D"}]} cf-ray: 9c00e6bfe9e63585-BLR alt-svc: h3=":443"; ma=86400 HTTP/2 302 date: Sun, 18 Jan 2026 20:39:23 GMT content-type: text/html;charset=utf-8 location: https://biorxiv.org/lookup/doi/10.1101/265231 server: cloudflare vary: Origin vary: Accept expires: Sun, 18 Jan 2026 21:11:00 GMT permissions-policy: interest-cohort=(),browsing-topics=() cf-cache-status: DYNAMIC nel: {"report_to":"cf-nel","success_fraction":0.0,"max_age":604800} strict-transport-security: max-age=31536000; includeSubDomains; preload report-to: {"group":"cf-nel","max_age":604800,"endpoints":[{"url":"https://a.nel.cloudflare.com/report/v4?s=Jq0AXtIIZPqLTYYHsRN8cAqJkIW9VSbEuci4BlLHlOzNgKUBd3nWxoZMhq8RFIhS9tH9STxSq5aTAFHHuz73IZL967csdg%3D%3D"}]} cf-ray: 9c00e6c15a893585-BLR alt-svc: h3=":443"; ma=86400 HTTP/1.1 302 Found Date: Sun, 18 Jan 2026 20:39:23 GMT Content-Type: text/html; charset=iso-8859-1 Transfer-Encoding: chunked Connection: keep-alive server: cloudflare location: https://www.biorxiv.org/lookup/doi/10.1101/265231 cf-cache-status: DYNAMIC Nel: {"report_to":"cf-nel","success_fraction":0.0,"max_age":604800} Report-To: {"group":"cf-nel","max_age":604800,"endpoints":[{"url":"https://a.nel.cloudflare.com/report/v4?s=rRLdc457cVCbTUZVYi%2B51rataaNiU439VFJ8fvmywxNMrZAUK3Fpna7duyaZf7PkGLOrYwTafv%2FMQN0JrR2fMVoQi8TjFrKLNxxI"}]} CF-RAY: 9c00e6c2fff2dab3-BOM alt-svc: h3=":443"; ma=86400 HTTP/2 301 date: Sun, 18 Jan 2026 20:39:24 GMT content-type: text/html; charset=UTF-8 location: https://www.biorxiv.org/content/10.1101/265231v2 cf-ray: 9c00e6c61e5eb22e-BLR x-content-type-options: nosniff x-content-type-options: nosniff x-drupal-cache: MISS expires: Sun, 18 Jan 2026 21:09:24 GMT cache-control: public, max-age=1800 x-varnish-ttl: pragma: no-cache vary: Accept-Encoding x-highwire-sitecode: biorxiv x-highwire-smart-code: biorxiv_production x-varnish: 697473691 x-varnish-cache: via: 1.1 varnish cf-cache-status: EXPIRED set-cookie: __cf_bm=F5mo3LNH2or5haJd4yF2b0PNi5YoSJKVE_W_4satJyc-1768768764-1.0.1.1-RGU0txvTJazayWdM7fpiP1Qz4iZB2tnD0g9RaWUrZMxVW0X9mMKyUl_ZUFu1hmK_R96Isqvdu4igVwzhSR7lw9pEpDXojGd4crfhl6Kmxl4; path=/; expires=Sun, 18-Jan-26 21:09:24 GMT; domain=.www.biorxiv.org; HttpOnly; Secure; SameSite=None server: cloudflare HTTP/2 200 date: Sun, 18 Jan 2026 20:39:26 GMT content-type: text/html; charset=utf-8 content-encoding: gzip x-content-type-options: nosniff x-content-type-options: nosniff x-drupal-cache: MISS expires: Sun, 19 Nov 1978 05:00:00 GMT cache-control: no-cache, must-revalidate set-cookie: SSESS1dd6867f1a1b90340f573dcdef3076bc=INuXmle7gSgqHL6Oi6mpEVODHhnDkeyo3cIqtj8F4u4; expires=Wed, 11-Feb-2026 00:12:44 GMT; path=/; domain=.biorxiv.org; secure; HttpOnly content-language: en x-frame-options: SAMEORIGIN x-generator: Drupal 7 (https://drupal.org) link: ; rel="canonical",; rel="shortlink" vary: Accept-Encoding x-highwire-sitecode: biorxiv x-highwire-smart-code: biorxiv_production x-varnish: 697473728 age: 0 via: 1.1 varnish x-varnish-ttl: x-varnish-cache: cf-cache-status: DYNAMIC server: cloudflare cf-ray: 9c00e6c9b829b22e-BLR End-to-end differentiable learning of protein structure | bioRxiv

Abstract

Predicting protein structure from sequence is a central challenge of biochemistry. Co‐evolution methods show promise, but an explicit sequence‐to‐structure map remains elusive. Advances in deep learning that replace complex, human‐designed pipelines with differentiable models optimized end‐to‐end suggest the potential benefits of similarly reformulating structure prediction. Here we report the first end‐to‐end differentiable model of protein structure. The model couples local and global protein structure via geometric units that optimize global geometry without violating local covalent chemistry. We test our model using two challenging tasks: predicting novel folds without co‐evolutionary data and predicting known folds without structural templates. In the first task the model achieves state‐of‐the‐art accuracy and in the second it comes within 1‐2Å; competing methods using co‐evolution and experimental templates have been refined over many years and it is likely that the differentiable approach has substantial room for further improvement, with applications ranging from drug discovery to protein design.

The copyright holder for this preprint is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.