March'24 Edition

March'24 Edition

Boltzmann Labs Revolutionizes Drug Discovery Suite with Powerful New Features!

Boltzmann Labs' AI-powered Drug Discovery suite is now even better at automating and streamlining the entire drug discovery process. We've significantly enhanced BoltChem, ReBolt, BoltPro, and BoltBio to deliver faster and less risky drug development. These optimized features translate to significant time and cost savings for researchers.

Ready to experience the power of AI in drug discovery?? Contact us today at [email protected] to learn more about our cutting-edge suite and try our industry-leading products.? Together, let's leverage the transformative power of AI to revolutionize drug development.



Novel RORγ Inhibitor: JTE-151 Shows Promise in Phase I Trial

Structure of JTE-151

RORγ, the Retinoic acid receptor-related Orphan Receptor gamma, plays a vital role in the development and function of Th17 cells, a specific type of immune cell critical for combatting infections. However, in autoimmune diseases, Th17 cells lose their way and begin attacking the body's own tissues. As RORγ acts as a central controller of Th17 cells, it becomes an appealing target. By inhibiting RORγ, researchers aim to decrease the production of inflammatory molecules like interleukin-17 (IL-17), This ability to target RORγ makes it a promising contender for the creation of new medications to address autoimmune disorders.

Scientists designed molecules to target RORγ, the key regulator of Th17 immune cells involved in autoimmune diseases.? JTE-151 emerged through a process optimizing drug-likeness and targeting specific regions of the RORγ protein.? The analysis confirmed JTE-151's binding to RORγ and its ability to suppress Th17 cells and IL-17 production in mice.

JTE-151, the first RORγ inhibitor made to be safe and effective, displayed encouraging outcomes in a phase I trial. Doses ranging from 30mg to 1600mg raised JTE-151 levels in the blood without any significant adverse effects, prompting the need for additional clinical investigations.

Reference: https://pubs.acs.org/doi/abs/10.1021/acs.jmedchem.8b01567?src=recsys


Molecular Glue Drug SJ3149 Shows Early Success in Preclinical Studies

The molecular “super-glue” SJ3149 connects CK1α (its target) to the protein that will mark that target for destruction.

Cancer researchers are exploring a novel approach to target specific proteins for destruction. Molecular glues, act like matchmakers in the cell, binding to unwanted proteins and the cell's natural disposal system and tagging them for destruction. This innovative approach targets proteins previously considered "undruggable" by conventional drugs. High-throughput screening (HTS) acts as a powerful tool in this process, rapidly sifting through vast libraries of candidate glues to find those that precisely fit the target protein and disposal system. This approach paves the way for the development of a new generation of cancer therapies, with molecular glue as the solution.

St. Jude children’s hospital researchers significantly improved a molecular glue compound (SJ7095) to create SJ3149, a "super-glue" targeting cancer-promoting protein CK1α.? This discovery came about through high-throughput screening (HTS) of a 3630-compound library against various childhood cancers.? HTS identified SJ3149, which after optimization showed potent and selective activity against CK1α in both in vitro and in vivo studies, leading to reduced tumor growth, all while exhibiting a favorable safety profile.

Boltchem leverages AI for structure-based molecule generation which enables them to design small molecules that bind to their targets with high affinity.

To know more about Boltchem book a demo at Boltchem

Reference:

https://oxfordglobal.com/discovery/resources/from-screening-to-success-how-molecular-super-glue-sj3149-could-advance-cancer-treatments?utm_campaign=Discovery%20Content&utm_content=286935073&utm_medium=social&utm_source=twitter&hss_channel=tw-374573556


AI Revolutionizes Protein Design: De novo Design of Alpha Proteins

Researchers have achieved a significant breakthrough in the design of complex proteins, departing from the traditional focus on simpler structures. This innovative method harnesses artificial intelligence (AI) to meticulously analyze existing protein structures, pinpointing common building blocks within them. By skillfully combining these fundamental units, scientists successfully engineered five distinct proteins featuring elaborate arrangements of alpha helices, a crucial component of protein structure. This newfound ability to craft intricate proteins marks a transformative advancement, as the functions of proteins are intricately tied to their 3D structures. With these more complex designs, the potential arises for the development of proteins with entirely novel functionalities, opening doors to diverse applications across healthcare and life sciences.

Central to this milestone achievement is the pivotal role of artificial intelligence (AI) in protein design. Delving into the vast repository of protein structures within the Protein Data Bank (PDB), the team identified 18 characteristic helix-loop-helix motifs. Through computational wizardry, they demonstrated the capability to generate a spectrum of all-α protein tertiary structures, ranging from simple to intricate, by amalgamating these typical motifs with canonical α-helices. This fusion of AI-driven design and protein engineering not only showcases the remarkable diversity of all-α protein structures but also heralds a new era of tailored protein creations with intricate shapes, offering immense potential for advancing healthcare and life sciences.

Reference: Design of complicated all-α protein structures | Nature Structural & Molecular Biology


BaseFold Achieves Breakthrough Accuracy in Predicting Complex Protein Shapes

With the emergence of AI models like AlphaFold2 that revolutionized protein structure prediction from slow methods like X-ray crystallography, a key challenge persists: these models rely on limited, biased public databases hindering their accuracy for complex proteins, especially those underrepresented in the data.

This is where Basecamp Research's BaseFold steps in. BaseFold tackles the challenge of achieving high-accuracy predictions for complex proteins by using a unique approach. Instead of relying solely on existing, limited databases, BaseFold leverages BaseGraph, a massive dataset built by Basecamp Research. BaseGraph is a treasure trove of genetic information collected through partnerships with over 25 biodiversity-rich countries. This dataset goes beyond just protein sequences; it includes rich genomic context and comprehensive metadata, providing a more complete picture of the protein world. By training AI models on BaseGraph, researchers observed significant improvements in their performance, including AlphaFold2.

Basecamp Research scientists compared BaseFold's performance with AlphaFold2 using protein structures from the CASP15 (Critical Assessment of Structure Prediction) competition and CAMEO (Continuous Automated Model Evaluation) project. The results were impressive. BaseFold, fueled by the rich data in BaseGraph, improved the accuracy of AlphaFold2's predictions by up to six times for complex protein structures, particularly those underrepresented in public databases. BaseFold not only predicts protein shapes better but also shows a 3x increase in accuracy for modeling drug-protein interactions, accelerating drug discovery.

BoltPro is an AI-powered protein engineering platform, that enables the generation of new protein sequences utilizing state-of-the-art Generative AI, predicting properties, and facilitating AI-guided mutagenesis for the creation of innovative and functional proteins. Try BoltPro?

Reference: IMPROVING ALPHAFOLD2 PERFORMANCE WITH A GLOBAL METAGENOMIC & BIOLOGICAL DATA SUPPLY CHAIN | bioRxiv


?BacPROTACs - A Novel Strategy? for Treating? ? ? Multidrug-Resistant TB

Researchers are exploring a promising new approach to fight antibiotic-resistant bacteria, including Mycobacterium tuberculosis (TB) which causes TB.? A new research utilizes proteolysis-targeting chimeras (PROTACs) to degrade bacterial proteins. PROTACs work by hijacking the bacteria's natural protein degradation machinery to target and eliminate specific proteins essential for bacterial survival.

Researchers focused on harnessing the ClpC1P1P2 protease system in mycobacteria. The researchers designed BacPROTACs, bifunctional molecules that link a bacterial protein targeted for degradation with a molecule that binds to the ClpC1 subunit of the protease system. They demonstrated that BacPROTACs could effectively degrade target proteins in Mycobacterium smegmatis and Mycobacterium tuberculosis, including strains resistant to multiple antibiotics.

BacPROTACs, a new class of antibiotics, cleverly hijack the bacteria's own protein disposal system to eliminate unwanted proteins. These bifunctional molecules act like bridges, with one end attaching to a specific protein targeted for destruction and the other end binding to a protein on the bacteria's ClpC1P1P2 protease. This creates a three-way complex, essentially tricking the protease into recognizing the target protein as flagged for disposal. The protease then breaks down the target protein, offering a potentially more specific and resistance-averse strategy to combat antibiotic-resistant bacteria.

This strategy offers a two-fold advantage, it can target proteins previously inaccessible to antibiotics, and the indirect degradation process makes it harder for bacteria to develop resistance. Early studies have shown effectiveness against multidrug-resistant TB strains, making BacPROTACs a promising avenue for developing new TB treatments, particularly against these challenging infections. Further research is needed to optimize PROTACs and advance them into clinical trials, but this new approach has the potential to be a game-changer in the fight against antibiotic resistance.

Boltchem, a small molecule discovery studio that leverages AI to design small molecules. Their approach utilizes generative algorithms trained on extensive experimental data. This allows them to not only design the core small molecules but also customize linkers and explore potential proteolysis-targeting chimeras (PROTACs). To know more about Boltchem book a demo at Boltchem

Reference: https://www.frontiersin.org/articles/10.3389/fchem.2024.1358539/full


要查看或添加评论,请登录

社区洞察

其他会员也浏览了