• Home
  • Julian M. Kleber
Julian M. Kleber

Julian M. Kleber
  • Master of Science
  • Developer at sail.black

I am a computational scientist. I build machine learning models, simulate molecules, and contribute to open-source tools

About

29
Publications
13,039
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
16
Citations
Current institution
sail.black
Current position
  • Developer
Additional affiliations
June 2022 - present
sail.black
Position
  • Developer
October 2020 - January 2021
sail.black
Position
  • PhD Student
Education
April 2020 - August 2022
Leuphana University of Lüneburg
Field of study
  • Sustainable Chemistry
October 2018 - March 2021
Freie Universität Berlin
Field of study
  • Computational Science
October 2018 - March 2021
Freie Universität Berlin
Field of study
  • Chemistry

Publications

Publications (29)
Preprint
Full-text available
Electrochemistry is in a renaissance and a new theoretical foundation to study electrochemical systems is needed. In this work, a novel method of describing electrochemical organic reactions is investigated computationally as well as experimentally. The work focuses on the study of ground and excited states during the heterogenous catalysis.
Preprint
Full-text available
The CACHE challenges are a series of prospective benchmarking exercises meant to evaluate progress in the field of computational hit-finding. Here we report the results of the inaugural CACHE #1 challenge in which 23 computational teams each selected up to 100 commercially available compounds that they predicted would bind to the WDR domain of the...
Thesis
Full-text available
Modelling of biodegradability is crucial for designing safe chemicals and directly addresses 8 of the 17 sustainable development goals (SDG). As for today, the modelling is mostly done via closed and proprietary software with lacking transparency development practices. The quality of the underlying data is questionable, as there is no central or op...
Thesis
Full-text available
For decades, the drug design community was concerned with deriving insights for their targets from a static picture using X-ray crystallography and pharmacophores. Likewise, computational methods for in silico studies of ligand binding rely heavily on docking, a static method. Still, computational methods such as molecular dynamics (MD) simulations...
Preprint
Full-text available
A central dogma is that fluorine is the most electro negative element in the PSE and would thusact as a pure electron acceptor. However, to some chemists it is known, that fluorine can bond induality. This means fluorine can act as a donor and an acceptor of bonds and thus electrons. Thedonor/acceptor properties of fluorine are known from the NF mo...
Preprint
Full-text available
A central dogma is that fluorine is the most electro negative element in the PSE and would thusact as a pure electron acceptor. However, to some chemists it is known, that fluorine can bond induality. This means fluorine can act as a donor and an acceptor of bonds and thus electrons. Thedonor/acceptor properties of fluorine are known from the NF mo...
Preprint
Full-text available
A central dogma is that fluorine is the most electro negative element in the PSE and would thus act as a pure electron acceptor. However, to some chemists it is known, that fluorine can bond in duality. This means fluorine can act as a donor and an acceptor of bonds and thus electrons. The donor/acceptor properties of fluorine are known from the NF...
Preprint
Full-text available
A bstract Computer-Aided Drug Design is advancing to a new era. Recent developments in statistical modelling, including Deep Learning , Machine Learning and high throughput simulations, enable workflows and deductions not achievable 20 years ago. The key interaction for many small molecules is via bio-molecules. The interaction between a small mole...
Research Proposal
Full-text available
Ziel Im Rahmen einer Masterarbeit und basierend auf öffentlich und wenn möglich auch noch nicht öffentlich zugänglichen Datensätzen möchte ich herausfinden, welche Moleküleigenschaften eine Rolle beim Bioabbau eine Rolle spielen. Hierfür sollen alle in der Literatur dokumentierten und in routinemäßig verwendeter Open-Source Software vorhandenen Des...
Preprint
Full-text available
This work investigates the possibility of quantum mechanical simulations through the standard MD algorithm using only the Lennard-Jones potential by simulating an overly dense Lennard-Jones fluid. The findings are verified against a FENE chain. The simulations suggest, that indeed a Bose-Einstein-Condesate was simulated and could offer a way to dir...
Presentation
Full-text available
Initial studies on how to make the lean synthesis pipeline reality and also on how to understand the role of iron in our liver.
Presentation
Full-text available
Our new pitch, 30.08.2021 today at Startup Incubator Berlin
Research Proposal
Full-text available
The simulation of protein-ligand interactions to solve questions for drug design is a vitally important step in drug design process. Many approaches such as docking are static approaches and if the drug designer considers dynamic information, classical MD simulations are considered. However, especially long-range, weak, di use, and complicated inte...
Preprint
Full-text available
Computer-Aided Drug Design is advancing to a new era. Recent developments in statistical modelling, including Deep Learning, Machine Learning and high throughput simulations enable workflows and deductions not achievable 20 years ago. The key interaction for many small molecules is via bio-molecules. The interaction between a small molecule and a b...
Research Proposal
Full-text available
Method
Full-text available
Initial study on highly experimental ligands, evaluating parts of the envisioned software pattern.
Poster
Full-text available
After being denied funding and some pre IP battles with the Freie Universität, I decided to da Hackathon for finding COVID-19 Leads. We targeted EGFR which might prove valuable for cancer therapies as well. Remember: We are Lean and can't do everything ten times and waste time and money. Let's hope someone synthesizes it and tests the 442. Hope t...
Research Proposal
Full-text available
Application for a research grant within the Berlin Research Alliance
Presentation
Full-text available
A presentation about the application of the synthesis schemas derived in biotransferations.
Method
Full-text available
The schema is a try to minimize efforts in antibiotic development and shows how to selectively address different carbonyl groups within polyketide bodies.
Research Proposal
Full-text available
Resarch on efficient multicyclic systems for drug development
Poster
Full-text available
Reaction scheme for overview
Preprint
Full-text available
It's summarising the reaction scheme and shows potential ways to go
Preprint
Full-text available
For a seminar, I assesed aceophenone in terms of green chemistry. Feedback welcome.

Questions

Questions (9)
Question
I am evaluating the options, especially in the true open-source world, for making reliable calculations with high-accuracy using quantum correlation methods.
Question
To store my research data, I am seeking options to digitally store scientific data reliably (immutable). What are your thoughts on it?
Question
I would love to hear your insights about the major differences between standard DNN and GNNs. What are the most important points in your opinion?
Question
Hello,
I was wondering if you could recommend any libraries for MD simulations of proteins with ligands in Python. They may also be accessible via bash.
They need to be available for free in any form and have completely open-sourced code.
Thanks
Question
Hello,
What are the best tools and workflows to do DFT inside an active pocket of a given protein and ligand?
Thanks
Question
Hello,
I am calculating a series of organic molecules containing elements from the 1st to the 5th period. My goal is to calculate the series with a minimum amount of time. However, only the relative ordering of the MOs is important. I am wondering how different basis sets influence the relative ordering of the MOs? Is there a one-solves-it-all basis set, or should I rather go with different basis sets for different molecules?
Thanks,
Julian
Question
Hello,
I have the feeling that everything in chemistry related to information and data in particular is very sharded and obscured. Are there any projects ensuring more transparency and openess of the data? I am getting quite lost to be honest and would wish for more standards and unity in the field.
Thanks,
Julian
Question
Hello,
I am building a database of simulated properties for chemicals. However, before starting, I was wondering if there is any research available on the efficiency of parallelisation of the ORCA quantum chemistry package. I am especially interested in the sweet spot of cores.
Thanks,
Julian
Question
Hello,
We are working on a project related to surface chemistry and spectroscopy. For this purpose, we want to build a database with chemical structures and related properties. We would prefer using open-source tools because the project is academic in nature. I found the combination of RDKit and PostgreSQL quite promising. What are your thoughts on the best database system for chemical structures?
Thanks,
Julian

Network

Cited By