-
Two-Stage Pretraining for Molecular Property Prediction in the Wild
Authors:
Kevin Tirta Wijaya,
Minghao Guo,
Michael Sun,
Hans-Peter Seidel,
Wojciech Matusik,
Vahid Babaei
Abstract:
Molecular deep learning models have achieved remarkable success in property prediction, but they often require large amounts of labeled data. The challenge is that, in real-world applications, labels are extremely scarce, as obtaining them through laboratory experimentation is both expensive and time-consuming. In this work, we introduce MoleVers, a versatile pretrained molecular model designed fo…
▽ More
Molecular deep learning models have achieved remarkable success in property prediction, but they often require large amounts of labeled data. The challenge is that, in real-world applications, labels are extremely scarce, as obtaining them through laboratory experimentation is both expensive and time-consuming. In this work, we introduce MoleVers, a versatile pretrained molecular model designed for various types of molecular property prediction in the wild, i.e., where experimentally-validated labels are scarce. MoleVers employs a two-stage pretraining strategy. In the first stage, it learns molecular representations from unlabeled data through masked atom prediction and extreme denoising, a novel task enabled by our newly introduced branching encoder architecture and dynamic noise scale sampling. In the second stage, the model refines these representations through predictions of auxiliary properties derived from computational methods, such as the density functional theory or large language models. Evaluation on 22 small, experimentally-validated datasets demonstrates that MoleVers achieves state-of-the-art performance, highlighting the effectiveness of its two-stage framework in producing generalizable molecular representations for diverse downstream properties.
△ Less
Submitted 18 July, 2025; v1 submitted 5 November, 2024;
originally announced November 2024.
-
TrustMol: Trustworthy Inverse Molecular Design via Alignment with Molecular Dynamics
Authors:
Kevin Tirta Wijaya,
Navid Ansari,
Hans-Peter Seidel,
Vahid Babaei
Abstract:
Data-driven generation of molecules with desired properties, also known as inverse molecular design (IMD), has attracted significant attention in recent years. Despite the significant progress in the accuracy and diversity of solutions, existing IMD methods lag behind in terms of trustworthiness. The root issue is that the design process of these methods is increasingly more implicit and indirect,…
▽ More
Data-driven generation of molecules with desired properties, also known as inverse molecular design (IMD), has attracted significant attention in recent years. Despite the significant progress in the accuracy and diversity of solutions, existing IMD methods lag behind in terms of trustworthiness. The root issue is that the design process of these methods is increasingly more implicit and indirect, and this process is also isolated from the native forward process (NFP), the ground-truth function that models the molecular dynamics. Following this insight, we propose TrustMol, an IMD method built to be trustworthy. For this purpose, TrustMol relies on a set of technical novelties including a new variational autoencoder network. Moreover, we propose a latent-property pairs acquisition method to effectively navigate the complexities of molecular latent optimization, a process that seems intuitive yet challenging due to the high-frequency and discontinuous nature of molecule space. TrustMol also integrates uncertainty-awareness into molecular latent optimization. These lead to improvements in both explainability and reliability of the IMD process. We validate the trustworthiness of TrustMol through a wide range of experiments.
△ Less
Submitted 26 February, 2024;
originally announced February 2024.
-
The Ion Fluorescence Chamber (IFC): A new concept for directional dark matter and topologically imaging neutrinoless double beta decay searches
Authors:
B. J. P. Jones,
F. W. Foss,
J. A. Asaadi,
E. D. Church,
J. deLeon,
E. Gramellini,
O. H. Seidel,
T. T. Vuong
Abstract:
We introduce a novel particle detection concept for large-volume, fine granularity particle detection: The Ion Fluorescence Chamber (IFC). In electronegative gases such as SF$_6$ and SeF$_6$, ionizing particles create ensembles of positive and negative ions. In the IFC, positive ions are drifted to a chemically active cathode where they react with a custom organic turn-on fluorescent monolayer enc…
▽ More
We introduce a novel particle detection concept for large-volume, fine granularity particle detection: The Ion Fluorescence Chamber (IFC). In electronegative gases such as SF$_6$ and SeF$_6$, ionizing particles create ensembles of positive and negative ions. In the IFC, positive ions are drifted to a chemically active cathode where they react with a custom organic turn-on fluorescent monolayer encoding a long-lived 2D image. The negative ions are sensed electrically with course resolution at the anode, inducing an optical microscope to travel to and scan the corresponding cathode location for the fluorescent image. This concept builds on technologies developed for barium tagging in neutrinoless double beta decay, combining the ultra-fine imaging capabilities of an emulsion detector with the monolithic sensing of a time projection chamber. The result is a high precision imaging detector over arbitrarily large volumes without the challenges of ballooning channel count or system complexity. After outlining the concept, we discuss R\&D to be undertaken to demonstrate it, and explore application to both directional dark matter searches in SF$_6$ and searches for neutrinoless double beta decay in large $^{82}$SeF$_6$ chambers.
△ Less
Submitted 18 March, 2022;
originally announced March 2022.
-
Precision and accuracy of single-molecule FRET measurements - a worldwide benchmark study
Authors:
Björn Hellenkamp,
Sonja Schmid,
Olga Doroshenko,
Oleg Opanasyuk,
Ralf Kühnemuth,
Soheila Rezaei Adariani,
Anders Barth,
Victoria Birkedal,
Mark E. Bowen,
Hongtao Chen,
Thorben Cordes,
Tobias Eilert,
Carel Fijen,
Markus Götz,
Giorgos Gouridis,
Enrico Gratton,
Taekjip Ha,
Christian A. Hanke,
Andreas Hartmann,
Jelle Hendrix,
Lasse L. Hildebrandt,
Johannes Hohlbein,
Christian G. Hübner,
Eleni Kallis,
Achillefs N. Kapanidis
, et al. (28 additional authors not shown)
Abstract:
Single-molecule Förster resonance energy transfer (smFRET) is increasingly being used to determine distances, structures, and dynamics of biomolecules in vitro and in vivo. However, generalized protocols and FRET standards ensuring both the reproducibility and accuracy of measuring FRET efficiencies are currently lacking. Here we report the results of a worldwide, comparative, blind study, in whic…
▽ More
Single-molecule Förster resonance energy transfer (smFRET) is increasingly being used to determine distances, structures, and dynamics of biomolecules in vitro and in vivo. However, generalized protocols and FRET standards ensuring both the reproducibility and accuracy of measuring FRET efficiencies are currently lacking. Here we report the results of a worldwide, comparative, blind study, in which 20 labs determined the FRET efficiencies of several dye-labeled DNA duplexes. Using a unified and straightforward method, we show that FRET efficiencies can be obtained with a standard deviation between $Δ$E = +-0.02 and +-0.05. We further suggest an experimental and computational procedure for converting FRET efficiencies into accurate distances. We discuss potential uncertainties in the experiment and the modelling. Our extensive quantitative assessment of intensity-based smFRET measurements and correction procedures serve as an essential step towards validation of distance networks with the ultimate aim to archive reliable structural models of biomolecular systems obtained by smFRET-based hybrid methods.
△ Less
Submitted 29 December, 2017; v1 submitted 10 October, 2017;
originally announced October 2017.