experiments | ARCHES Initiative

Journal Club: April 2020

By meb16 | Published April 23, 2020

Scientists have turned the structure of the coronavirus into music

Thursday, April 23, 2020

Our third journal club is online! Feel free to join the discussion remotely in the comments below. One of our participants in Project CHROMA suggested the following research that combines science and music to study the coronavirus:

Vineeth Venugopal (2020) “Scientists have turned the structure of the coronavirus into music,” Science 10.1126/science.abc0657.
https://www.sciencemag.org/news/2020/04/scientists-have-turned-structure-coronavirus-music

Researchers at MIT have created a musical representation of the amino acid sequence and structure of the COVID-19 spike protein (based on protein data bank entry 6VSB published by Wrapp et al. 2020) using a technique known as sonification.

The team led by Markus Buehler transposed the vibrational frequencies of the 20 natural amino acids to an audible spectrum in order to assign a musical note to each amino acid, thus creating a 20-tone “amino acid scale.” To create the COVID-19 spike protein musical score, the notes were played on a Japanese koto. The volume and duration of each note were defined by the secondary and higher-order folded structure of the protein. Heat-induced molecular vibrations were represented by unique sounds as well. A neural network was then used to generate music compositions that captured the relationships between amino acid sequence and protein structure.

It’s apparently faster using this technique rather than traditional molecular dynamics modeling to search for sites where antibodies or drugs could bind on the viral protein — researchers simply have to compare the musical scores of the sonified structures.

This musical technique is also a great way to communicate the significance of protein sequences and their folded structure to the public!

Read more about this sonification method developed by the research team:

Chi-Hua Yu, Zhao Qin, Francisco J Martin-Martinez, and Markus J Buehler (2019) “A Self-Consistent Sonification Method to Translate Amino Acid Sequences into Musical Compositions and Application in Protein Design Using Artificial Intelligence,” ACS Nano, 13(7): 7471-7482.
https://pubs.acs.org/doi/10.1021/acsnano.9b02180

There’s also an interactive phone app available for download on the Google Play Store: https://play.google.com/store/apps/details?id=com.synth.aminoacidplayer

Posted in COVID-19, Events, Journal Club | Comments Off

Journal Club: March 2020

By meb16 | Published March 19, 2020

Systematically Improving Espresso: Insights from Mathematical Modeling and Experiment

Thursday, March 19, 2020

Our second journal club was moved online! Feel free to join the discussion remotely in the comments below. Josh Hill from the department of Physics & Astronomy suggested the following paper that combines science and the culinary arts:

Michael Cameron, Dechen Morisco, Daniel Hofstetter, Erol Uman, Justin Wilkinson, Zachary Kennedy, Sean Fontenot, William Lee, Christopher Hendon, and Jamie M. Foster (2020) “Systematically Improving Espresso: Insights from Mathematical Modeling and Experiment,” Matter 2(3):631-648.
https://www.sciencedirect.com/science/article/pii/S2590238519304102

The coffee industry is huge! In the U.S. alone, it provided 1.5 million jobs and accounted for 1.6% of the gross domestic profit in 2015. Espresso is the trickiest coffee beverage format in terms of maintaining a consistent yield and desirable flavor. The goal of this project was to develop a mathematical model of espresso extraction in order to optimize the espresso parameters at play for a reduction of the variation in taste and waste!

These espresso parameters are:

grind setting
coffee mass
water pressure and temperature
beverage volume

The model determined that variation in espresso drinks was not due to human variation, but rather due to non-uniform flow during espresso extraction related to coffee grind size. The team determined a critical minimum grind size that would homogenize the flow, increase extraction yield, and reduce drink variability. Interestingly, the extraction yield had a non-linear dependence on grind setting — this was attributed to a competing relationship between a finer grind leading to an increase in flow but also an increase in grind aggregation leading to partial clogging conditions that affect the flow.

Typically, 20g of dry coffee mass is used to make a single 40g espresso shot. However, the research team demonstrated that a barista can achieve highly reproducible espresso with the same 40g extraction yield by reducing the coffee mass to 15g and using a counter-intuitively coarser grind!

The mass reduction suggestions were implemented at a local cafe in Eugene, Oregon. Espresso drinks were prepared with 15g of specialty-grade coffee, rather than 20g. Firstly, there was a reduced order-to-delivery time since the shot brewing time was 14s rather than 20-30s. Secondly, the research team calculated that the cafe had saved $0.13 per drink given the lower coffee dry mass being used. This amounted to an increase in profit of $3,620 per year!

To summarize: The multi-scale mathematical model of espresso extraction enabled an understanding of the origin of the variation in espresso drinks. The research team made suggestions to minimize drink variation and dry coffee waste: (1) reduce the dry coffee mass used and (2) increase the grind size. Espresso yield and taste were both preserved, while the shot brewing time became faster.

These novel, model-based brewing protocols will contribute to creating a more sustainable coffee-consuming future!

All posts