Data-Driven Organic Materials Lab

Seifrid research group in MSE at NC State

[Data-Driven Organic Materials Lab @ddom_lab]

We think there are important implications for anyone using ML to guide #polymer design or build a #SelfDrivingLab.

[Data-Driven Organic Materials Lab @ddom_lab]

Out now in @aplmachinelearn: the group's first-ever publication!! 🥳

https://t.co/CM0CuhjtrP

[Data-Driven Organic Materials Lab @ddom_lab]

Good in-distribution performance is the easy part. The harder question is when and why models fail on new polymers and formulations, and the answer isn't as simple as "more data."

[Martin Seifrid @M_Seifrid]

Not to toot my own horn, but I think this is incredibly important and a very difficult challenge to overcome quickly. Hopefully our call to action can spur faster progress!

[Data-Driven Organic Materials Lab @ddom_lab]

We explore how model performance can be assessed in this scenario, and use conjugated polymer conformation as a test case.

[Data-Driven Organic Materials Lab @ddom_lab]

🚨🚨 The group's first preprint is up on
@ChemRxiv! 🥳🥳

Robust Learning from Literature Data: Model Generalizability and Uncertainty for Predicting Conjugated Polymer Solution Conformation

📝 https://t.co/H1uKb8RF5a

[Data-Driven Organic Materials Lab @ddom_lab]

Conventional or #selfdrivinglab experiments can be informed by data gathered from the literature.

Important scientific challenges often require the development of previously unknown materials: materials discovery.

#machinelearning models are not designed for this scenario...

[Martin Seifrid @M_Seifrid]

Do you know of any papers that have datasets of organic materials (polymers, molecules) with experimental parameters–not material properties, but stuff like concentration, temperature, whatever?

[Research Corporation for Science Advancement @RCSA1]

Congrats to #Scialog AUT awardees Mark Hendricks @whitmancollege, Jessica Sampson @jesstheligand @UDResearch @ChemistryUD & Martin Seifrid @M_Seifrid @ddom_lab @ncstatemse @NCStateEngr @NCState!

[Data-Driven Organic Materials Lab @ddom_lab]

Another instrument is up and running! We set up our @GyrosProteinTec PurePep Chorus last Friday. Keep an eye out for exciting developments on this front 👀

[Data-Driven Organic Materials Lab @ddom_lab]

We just completed the installation of Big Purchase #1™️: a quadruple (!) detector SEC from @TosohBio. This bad boy has RI, UV, MALS, & viscometry 💪 We're really excited to start getting some absolute molecular weights!

[Research Corporation for Science Advancement @RCSA1]

RCSA, @BeckmanFnd & Frederick Gardner Cottrell Foundation announce awards to 7 multidisciplinary teams in the 1st year of #Scialog: Automating Chemical Laboratories, which aims to accelerate innovation & broaden access within the chemical enterprise. https://t.co/KYPFDXokQm

[Research Corporation for Science Advancement @RCSA1]

Congratulations to @M_Seifrid @ddom_lab @ncstatemse @NCStateEngr & @CoryMSimon @EngineeringOSU & @cgbischak @UtahChemistry @uofu_science for “Reducing the Cost of Device Development with Closed-Loop Proxy Measurements and Supplemental Characterization.”

[Martin Seifrid @M_Seifrid]

Our exploration of machine learning to predict #OPV device performance from molecular structure of the materials *and* processing data is now officially published in @JMaterChem A as part of their 2024 Emerging Investigators series!

https://t.co/dve9E2jvkz

[Data-Driven Organic Materials Lab @ddom_lab]

@ncstatemse @NCStateEngr @opentrons And it's only the 25th one that's been installed so far

[Data-Driven Organic Materials Lab @ddom_lab]

First semester at @ncstatemse is in the books! Not only is @NCStateEngr building the future of #SelfDrivingLabs, but there's an awesome self-driving library

[Data-Driven Organic Materials Lab @ddom_lab]

@ncstatemse @NCStateEngr The semester's big achievement was installing our new @opentrons Flex