Macrocyclic peptides are an emerging modality that combines biologics-like affinity with small-molecule-like developability, but their vast combinatorial space and multi-parameter objectives make lead optimization slow and challenging. Prior generative approaches such as PepINVENT require chemists to pre-specify mutable positions for optimization, choices that are not always known a priori, and rely on static pretraining and optimization algorithms that limit the model's ability to generalize and effectively optimize peptide sequences. We introduce PepEVOLVE, a position-aware, dynamic framework that learns both where to edit and how to dynamically optimize peptides for multi-objective improvement. PepEVOLVE (i) augments pretraining with dynamic masking and CHUCKLES shifting to improve generalization, (ii) uses a context-free multi-armed bandit router that discovers high-reward residues, and (iii) couples a novel evolving optimization algorithm with group-relative advantage to stabilize reinforcement updates. During in silico evaluations, the router policy reliably learns and concentrates probability on chemically meaningful sites that influence the peptide's properties. On a therapeutically motivated Rev-binding macrocycle benchmark, PepEVOLVE outperformed PepINVENT by reaching higher mean scores (approximately 0.8 vs. 0.6), achieving best candidates with a score of 0.95 (vs. 0.87), and converging in fewer steps under the task of optimizing permeability and lipophilicity with structural constraints. Overall, PepEVOLVE offers a practical, reproducible path to peptide lead optimization when optimal edit sites are unknown, enabling more efficient exploration and improving design quality across multiple objectives.

Designing therapeutic peptides with tailored properties is hindered by the vastness of sequence space, limited experimental data, and poor interpretability of current generative models. To address these challenges, we introduce PepThink-R1, a generative framework that integrates large language models (LLMs) with chain-of-thought (CoT) supervised fine-tuning and reinforcement learning (RL). Unlike prior approaches, PepThink-R1 explicitly reasons about monomer-level modifications during sequence generation, enabling interpretable design choices while optimizing for multiple pharmacological properties. Guided by a tailored reward function balancing chemical validity and property improvements, the model autonomously explores diverse sequence variants. We demonstrate that PepThink-R1 generates cyclic peptides with significantly enhanced lipophilicity, stability, and exposure, outperforming existing general LLMs (e.g., GPT-5) and domain-specific baseline in both optimization success and interpretability. To our knowledge, this is the first LLM-based peptide design framework that combines explicit reasoning with RL-driven property control, marking a step toward reliable and transparent peptide optimization for therapeutic discovery.

This study uses double/debiased machine learning (DML) to evaluate the impact of transitioning from lecture-based blended teaching to a flipped classroom concept. Our findings indicate effects on students' self-conception, procrastination, and enjoyment. We do not find significant positive effects on exam scores, passing rates, or knowledge retention. This can be explained by the insufficient use of the instructional approach that we can identify with uniquely detailed usage data and highlights the need for additional teaching strategies. Methodologically, we propose a powerful DML approach that acknowledges the latent structure inherent in Likert scale variables and, hence, aligns with psychometric principles.

Tupac's humiliating intimate disfigurement revealed... and how his lies to cover it up led to his murder I've started having heart palpitations. 'Black Ivy League' university looks to expand into crime-riddled Oakland Kristen Bell's friends turn on her with savage disclosures: Insiders reveal poisonous whispers behind her back... as she goes into full diva mode Shooting leaves two dead and 11 injured at large house party with'underage people' in North Carolina Kim Kardashian's just been caught in a despicable lie. She can cry all she wants... there's no hiding the truth now: CAROLINE BULLOCK The'marry me' sex move that'll make even the most commitment-phobic of men beg to see you again... and it worked for THREE of my friends Prosecutor who declined to charge Letitia James with bank fraud fired after'mishandling evidence' Californians being urged to take up arms to deal with'aggressive' invasive species attacking children Inside Andrew's family summit: How Fergie wailed and'melted down' at title loss, Beatrice and Eugenie were'blindsided' and now daughters' assets face'ethics check' to avoid more scandal: BARBARA DAVIES LIZ JONES: I was devastated when my husband cheated. But here's the reason part of me was secretly glad that every woman over-50 will understand Psychotherapist explains why No Kings rallies consisted of mostly'educated white women' Tree optical illusion messes with your mind - you can see the squirrel but can you spot the cat in 30 seconds? Turn off the lights, burn a candle, look into the mirror and say the magic words: 'Bloody Mary, Bloody Mary, Bloody Mary'.

Stunning before-and-after photos show the seven most dramatic changes in Trump's controversial White House makeover She was a respected Teacher of the Year finalist... until she lost everything when Charlie Kirk was killed. Inside Andrew's family summit: How Fergie wailed and'melted down' at title loss, Beatrice and Eugenie were'blindsided' and now daughters' assets face'ethics check' to avoid more scandal: BARBARA DAVIES I have no sympathy for Britney Spears. What if her latest stunt had killed a kid? It's time to admit the truth about this public menace: KENNEDY'Nazi texts' leakers UNMASKED: Alleged White House saboteurs are finally exposed... and so is their twisted motive for destroying political prodigy Extraordinary story behind GM's decision to ax much-loved CarPlay... and sinister reason ALL manufacturers will follow What is Charcot-Marie-Tooth disease... the devastating condition that killed 9-1-1 Nashville actor Isabelle Tate Bijou Phillips files to change daughter's name after ex Danny Masterson's rape conviction Treasure hunters seeking Nazi gold worth £200MILLION believe they have'found the real thing' after'monumental' discovery under remains of SS palace'brothel' Former Gambino mob boss'Sammy the Bull' Gravano reveals the truth behind the NBA betting scandal My wife won't get a job and I feel broken trying to provide for our family. Hold on, says DEAR CAROLINE... that's bad enough but your letter raises a MUCH bigger red flag I got the body of my dreams at 51 by following 9 simple rules, says beauty guru ROSIE GREEN.

Billionaire Illinois Democrat governor caught in lie live on Fox News while trying to downplay Chicago's murder capital status Storm Melissa to explode into Category 5 hurricane as models reveal its'life-threatening' path to the US JAN MOIR: The Queen was blindly devoted to Prince Andrew... she raised a monster. The final hours of chess grandmaster Daniel Naroditsky - friends' desperate attempts to save him, warnings in final monologue and how he was haunted by sinister figure in hidden underworld. My wife won't get a job and I feel broken trying to provide for our family. Hold on, says DEAR CAROLINE... that's bad enough but your letter raises a MUCH bigger red flag Wild resurfaced Gilbert Arenas'snitching' claim goes viral in the wake of NBA mafia gambling scandal Inside the nondescript Virginia warehouse that wiped out the internet with one outage... and the neighbors who warn the next one is just a matter of time Fury as'insane' GM kills much-loved feature from upcoming cars as rival Ford doubles down I know all the secrets of the NBA legends' betting scandal. I think I've discovered Meghan's secret plan for if - or when - William strips away the Sussexes' royal titles: SHARON HUNT Disney fans left devastated after theme park dramatically'scales back' on its villains Doctor's $1M show of loyalty for murderer husband after he let adorable daughter, 2, die in roasting car as he watched adult videos Storm Melissa to explode into Category 5 hurricane as models reveal its'life-threatening' path to the US Tropical Storm Melissa is expected to strengthen into a life-threatening Category 5 hurricane that could swerve into the northeastern US in just days.

Tearful Kim Kardashian, 45, reveals doctors found brain aneurysm after MRI... as she blames stressful Kanye West divorce As royal insiders dish the dirt, this is what I'm told is the truth about Prince Andrew's daughters This is the exact plan I followed to supercharge my weight loss... and the surprising jab side-effect that cured me of my REAL problem: SUSAN ANDERSON Finance guru storms out of podcast with illegal migrants $420K in debt who insist they'deserve' new car and pool Dakota Johnson reveals her biggest'red flag' in men after Chris Martin split'Gaslighting' and'black out' fights: Kristen Bell and Dax Shepard's'volatile' marriage laid bare by insiders The secret calls and frantic meetings over Congressman's alleged affair with aide who set herself on fire in scandal that could upend Trump's future Pete Hegseth dealt another blow as judge shoots down effort to rebrand Pentagon with'warrior ethos' There's a taboo most men find repulsive... but if they can handle it, says JANA HOCKING, it's the biggest turn on ever The real reason behind Cracker Barrel's disastrous logo change... and it makes complete sense Astonishing new video shows Louvre robbers escaping in a mechanical delivery basket with £76m-worth of jewels - after evading CCTV that was'pointing the wrong way' Elon Musk's ex Grimes baffles fans with bizarre circular face tattoo as they insist inking looks like RINGWORM Putin ally accuses Trump of an'act of war' against Russia after US president imposed new oil sanctions French girl Lola, 12, who was'raped and murdered by Algerian woman' begged'please don't hurt me' before she was brutally killed, court hears Dave Grohl on'thin ice' with wife Jordyn Blum as insiders reveal her strict list of rules to save their marriage... and his plans for daughters to build relationship with his love child Amazon's delivery drivers will be forced to wear AI GLASSES that give them turn-by-turn directions to shave seconds off deliveries READ MORE: Amazon workers claim'kill switch' triggered massive outage In a bid to shave seconds off deliveries, Amazon will soon force its delivery drivers to wear smart glasses. The futuristic glasses use artificial intelligence ( AI) to feed drivers turn-by-turn directions leading up to customers' doorsteps. They're also fitted with cameras so drivers can scan packages and capture proof of delivery. Amazon claims the dystopian device will make deliveries'as safe and seamless as possible'. However, it seems not everyone agrees.

Chemical pretrained models, sometimes referred to as foundation models, are receiving considerable interest for drug discovery applications. The general chemical knowledge extracted from self-supervised training has the potential to improve predictions for critical drug discovery endpoints, including on-target potency and ADMET properties. Multi-task learning has previously been successfully leveraged to improve predictive models. Here, we show that enabling multitasking in finetuning of chemical pretrained graph neural network models such as Kinetic GROVER Multi-Task (KERMT), an enhanced version of the GROVER model, and Knowledge-guided Pre-training of Graph Transformer (KGPT) significantly improves performance over non-pretrained graph neural network models. Surprisingly, we find that the performance improvement from finetuning KERMT in a multitask manner is most significant at larger data sizes. Additionally, we publish two multitask ADMET data splits to enable more accurate benchmarking of multitask deep learning methods for drug property prediction. Finally, we provide an accelerated implementation of the KERMT model on GitHub, unlocking large-scale pretraining, finetuning, and inference in industrial drug discovery workflows.

We show that TKRR can achieve a fast rate of convergence in the over-aligned regime.