HOUSTON, Feb. 16, 2024 Rice University computer science researchers have found bias in widely used machine learning tools used for immunotherapy research.

Ph.D. students Anja Conev, Romanos Fasoulis and Sarah Hall-Swan, working with computer science faculty members Rodrigo Ferreira and Lydia Kavraki, reviewed publicly available peptide-HLA (pHLA) binding prediction data and found it to be skewed toward higher-income communities. Their paper examines the way that biased data input affects the algorithmic recommendations being used in important immunotherapy research.

Peptide-HLA Binding Prediction, Machine Learning and Immunotherapy

HLA is a gene in all humans that encodes proteins working as part of our immune response. Those proteins bind with protein chunks called peptides in our cells and mark our infected cells for the bodys immune system, so it can respond and, ideally, eliminate the threat.

Different people have slightly different variants in genes, called alleles. Current immunotherapy research is exploring ways to identify peptides that can more effectively bind with the HLA alleles of the patient.

The end result, eventually, could be custom and highly effective immunotherapies. That is why one of the most critical steps is to accurately predict which peptides will bind with which alleles. The greater the accuracy, the better the potential efficacy of the therapy.

But calculating how effectively a peptide will bind to the HLA allele takes a lot of work, which is why machine learning tools are being used to predict binding. This is where Rices team found a problem: The data used to train those models appears to geographically favor higher-income communities.

Why is this an issue? Without being able to account for genetic data from lower-income communities, future immunotherapies developed for them may not be as effective.

Each and every one of us has different HLAs that they express, and those HLAs vary between different populations, Fasoulis said. Given that machine learning is used to identify potential peptide candidates for immunotherapies, if you basically have biased machine models, then those therapeutics wont work equally for everyone in every population.

Redefining Pan-Allele Binding Predictors

Regardless of the application, machine learning models are only as good as the data you feed them. A bias in the data, even an unconscious one, can affect the conclusions made by the algorithm.

Machine learning models currently being used for pHLA binding prediction assert that they can extrapolate for allele data not present in the dataset those models were trained on, calling themselves pan-allele or all-allele. The Rice teams findings call that into question.

What we are trying to show here and kind of debunk is the idea of the pan-allele machine learning predictors, Conev said. We wanted to see if they really worked for the data that is not in the datasets, which is the data from lower-income populations.

Fasoulis and Conevs group tested publicly available data on pHLA binding prediction, and their findings supported their hypothesis that a bias in the data was creating an accompanying bias in the algorithm. The team hopes that by bringing this discrepancy to the attention of the research community, a truly pan-allele method of predicting pHLA binding can be developed.

Ferreira, faculty advisor and paper co-author, explained that the problem of bias in machine learning cant be addressed unless researchers think about their data in a social context. From a certain perspective, datasets may appear as simply incomplete, but making connections between what is or what is not represented in the dataset and underlying historical and economic factors affecting the populations from which data was collected is key to identifying bias.

Researchers using machine learning models sometimes innocently assume that these models may appropriately represent a global population, Ferreira said, but our research points to the significance of when this is not the case. He added that even though the databases we studied contain information from people in multiple regions of the world, that does not make them universal. What our research found was a correlation between the socioeconomic standing of certain populations and how well they were represented in the databases or not.

Professor Kavraki echoed this sentiment, emphasizing how important it is that tools used in clinical work be accurate and honest about any shortcomings they may have.

Our study of pHLA binding is in the context of personalized immunotherapies for cancer a project done in collaboration with MD Anderson, Kavraki said. The tools developed eventually make their way to clinical pipelines. We need to understand the biases that may exist in these tools. Our work also aims to alert the research community on the difficulties of obtaining unbiased datasets.

Conev noted that, though biased, the fact that the data was publicly available for her team to review was a good start. The team is hoping its findings will lead new research in a positive direction one that includes and helps people across demographic lines.

Ferreira is an assistant teaching professor of computer science. Kavraki is the Noah Harding Professor of Computer Science, a professor of bioengineering, electrical and computer engineering and director of the Ken Kennedy Institute for Information Technology.

The research was supported by the National Institutes of Health (U01CA258512) and Rice University.

Source: John Bogna, Rice University

See the rest here:

Rice University Researchers Uncover Bias in Machine Learning Tools for Immunotherapy - Datanami

(L-R) Sudheer Doss, PhD; Jack DiGiovanna, PhD; and Kawther Abdilleh, PhD.

February 11 is International Day of Women and Girls in Science. The day is an opportunity to promote full and equal access to and participation in science for women and girls and highlight the challenges they still face in the field.

Women in science, technology, engineering and mathematics (STEM) are extremely underrepresented. According to the Global Gender Gap Report (2023), women only comprise of a little over 29% of the field. The world needs more women in science because women in science change the world.

To celebrate this important day, weve asked PanCANs Director of Data Science and Informatics Kawther Abdilleh, PhD, six questions about women in STEM and what and who inspired her to choose a career in the sciences.

PanCAN: When did you realize you wanted to go into STEM as a career?

Kawther: At a young age, I was introduced to biology, chemistry, and computer science by my middle school teachers as well as my aunts and uncles, who were all pursuing careers in STEM disciplines. This led to my desire to follow in their footsteps and pursue a career in STEM.

PanCAN: What did your educational and career path look like how did you end up at PanCAN?

Kawther: I have a BS degree in Computer Science from Marymount University and a PhD in Molecular & Cellular Biology from the University of Maryland, College Park. After graduate school, I worked as a postdoctoral fellow at Georgia Tech. Throughout my academic experiences, I used bioinformatics and computational tools to analyze large-scale genomics data.

Before joining PanCAN, I worked on teams where I helped design and build bioinformatics data solutions for cancer researchers.

I came across a job advertisement for a position to manage the PanCAN SPARK platform. I was eager to apply all the skills Ive acquired over the years to this position at PanCAN.

PanCAN: What is PanCAN doing that excites you the most?

Kawther: I am most excited about PanCANs research and clinical initiatives. I am impressed with how PanCAN continues to move the needle forward in the fight against pancreatic cancer. Not only is PanCAN funding leading-edge pancreatic cancer research, but it is also leading in research and clinical efforts to improve patient outcomes in programs, such as the Early Detection Initiative, Precision Promise, and the SPARK platform.

PanCAN: What are some ways we can encourage women to embrace science early on?

Kawther: I think two main ways we can encourage young girls and women to embrace science early on are:

PanCAN: How important is representation in science?

Kawther: Representation is very important in science. Having relatable mentors is a key motivator for underrepresented groups to enter scientific fields. It really brings diversity of thought and experiences to science which contributes to the richness of the field in the form of new ideas and potentially new solutions.

PanCAN: Do you have any female mentors or role models in your field?

Kawther: Two of my aunts worked in research labs in biopharma. As a kid, I loved hearing about the research they worked on and was always interested in learning more. They helped foster my love of science and encouraged me to pursue a career in STEM.

Want to learn more about Kawther and the work shes doing at PanCAN? Watch her talk about SPARK, PanCANs health data integration platform for pancreatic cancer research:

A big PanCAN thank you to Kawther and women around the world making a difference in the STEM field.

Follow this link:

Women in STEM: 6 Questions With PanCAN's Kawther Abdilleh, PhD - Pancreatic Cancer News & Stories

Considering the rise of AI, along with recent layoffs by companies such as Google, Amazon, and Microsoft, you might be wondering if a career in the technology sector is a smart move. Despite some of these challenges, jobs like coding and programming are still rewarding and worthwhile. In fact, the average annual wage for a computer programmer is $102,790, according to the Bureau of Labor Statistics.

However, if youre interested in pursuing a career in this field, you might be wondering: What is the difference between coding and programming, exactly?

ADVERTISEMENT

Coding and programming are often used as synonyms when, in fact, they are not, said Dimitry Graf, engineering and program management leader at the software company, Canonical. He explained that when you start a career in software development, you typically begin with coding. That includes writing computer code, testing, and debugging it.

As you progress in your career, you naturally transition to programminga more high-level, end-to-end activity that involves planning, designing, deploying, maintaining, and scaling on top of writing and debugging code, Graf said.

He added that programmers interact more with business departments (marketing, sales, product, etc.) than coders do. This shifts the focus from tech skills to soft skills such as communication, project management, analytical skills, and leadership, he noted.

If youre considering a career in this field, learn more about the difference between coding and programmingand which one might be right for you.

Coding is the process of translating logic and requirements into a computer-readable language. The primary focus is on writing lines of code in a specific language (like Python, Java, or HTML) to perform a specific task or solve a particular problem.

Programming is a more comprehensive process that may include coding, but it also includes the planning, structuring, testing, and maintenance of software applications and systems. It involves understanding and implementing algorithms, data structures, and system design principles.

In terms of careers, coding and programming are closely related and often overlap, but there are some important distinctions, too.

Coding usually refers to the process of writing code in a programming language to make a computer perform a specific task. In other words, coding is a specific task within the realm of programming that focuses on writing actual lines of code, according to Edward Kim, vice president of education and training at Code Ninjas.

Programming, on the other hand, involves not only writing code but also the planning, design, testing, and maintenance of software. Consider programming as a broader superset of activities during the lifecycle of developing software in order to implement a full solution to address specific or an array of challenges and problem sets or needs, Kim said.

Coders rely heavily on text editors such as Sublime Text, Atom, and Visual Studio Code, which are all popular because of their simplicity, speed, and flexibility. They can also be enhanced with plugins. You might also use code libraries such as jQuery or React to simplify certain tasks.

In programming, its common to use frameworks such as Angular, .NET, or Ruby on Rails, which provide a more structured, standardized way to build applications.

Additionally, coding involves syntax-specific tools, like linters and compilers (ESLint for JavaScript or PEP 8 for Python, for example). Programming involves additional project management and collaboration tools such as Git or JIRA to manage the software development lifecycle.

For coding, you need a strong understanding of one or more programming languages and their syntax.

For programming, you not only need to have deep knowledge of the language, but also be able to think holistically about how the software works and how itll be used. It involves multiple categories of tasks to complete: Problem solving, algorithm design, overall architecture and organization of backend processes, implementation of specific code, testing and optimization, and quality control, Kim said.

In terms of difficulty, it can vary depending on your particular skills and interests. Some people might find the logic and problem-solving aspects of programming more challenging, while others might find the syntax and specifics of coding more difficult. Generally, programming is considered to encompass more complex and diverse tasks than coding, which is just one aspect of programming.

Its actually common for people to start out as coders and then transition into programming. Employers expect you to progress from a coder to a programmer, learn new skills, and take on more responsibilities, Graf said. That doesnt mean you have to build a vertical career, though. Not everyone wants and needs to be a manager, and thats fine. However, he reiterated, companies do expect people in software engineering to constantly upskill.

Both coding and programming are excellent fields to pursue as careers. They offer high demand, good earning potential, and opportunities for growth. However, while coding and programming are closely related, they have distinct differences in the realm of software development.

Coding is integral to creating software, but its just one part of a broader landscape. Programming, on the other hand, encompasses a wider scope. As a programmer, you not only need to understand coding, but also be skilled at problem-solving, logical thinking, and software architecture and design.

Whether you decide to pursue coding or programming will depend on your interests, skills, and career goals. But many people start out as coders and then progress to programming or another higher-level position as they gain more experience and skills.

More here:

Coding vs. programming: Whats the difference? - Fortune

By Nazrul Khandaker

Jadyn Worthington, now a senior majoring in computer science at University of Delaware, attended the York College-hosted NASA STEM Program at the age of 10 and acknowledged the impact the program had on him at a young age with great sincerity.

Beyond excelling in academia, Jadyn is a trailblazer utilizing his expertise in Machine Learning (ML) to inspire and educate the next generation. Jadyn keeps quite busy, and is involved in various campus activities and pioneering educational programs for elementary school kids, introducing them to the wonders of ML and paving the way for many to opt in to further STEM learning.

The NASA-SEMAA K1-12 (Science, Engineering, Mathematics and Aerospace Academy) program was introduced to the greater York community in 1999 with the help of Congressman Meeks to foster STEM education, engage underrepresented community children in hands-on learning activities, and create a pathway for involved students to plan on undertaking STEM courses and eventually become scientists. The SEMAA program ended in 2015 and was succeeded by MUREP (Minority University Research and Education Project) Aerospace Academy. York administered both SEMAA and MAA and graduated over 30,000 students over the years.

Like Jadyn, many NASA STEM graduates from York College chose to pursue STEM majors and as a result, became STEM graduates and professionals as well. Saintedym Sandy Wills, a six grade SEMAA participant in 2000, became the programs first-known PhD degree recipient from Duke University in 2018. An immunologist, Dr. Saintedym Wills is the Technical Director for Immunology Research and Development at Laboratory Corporation of America.

York not only offered STEM education to the children, it also tracked students once they left the program. The longitudinal tracking of NASA STEM graduates often brought to light inspiring and exemplary reflections from these tracked students, and York made it an effort to share these with the greater York Community on a regular basis. These graduates are stars and their success stories need to be conveyed in a manner where incoming generations of youths can follow their footsteps, draw inspiration, and participate in an active STEM learning community.

The program director Nazrul Khandaker reached out to Jadyn recently and received a marvelous response to share with the community children:

By Jadyn Worthington

The curriculum, meticulously designed by NASA, introduced me to a world of inquiry-based learning. Each session was a gateway to understanding national science, math, and technology standards, connected profoundly to NASA's mission directorates. Through engaging in projects and activities such as the Summer of Innovation, Engineering Design Challenges, learning mathematics, and learning information about planets in our solar system, I discovered the thrill of applying theoretical knowledge to practical endeavors while doing some of the hands-on work.

Motivation and Impact: The guidance and mentorship provided by Dr. Nazrul Khandaker within the NASA-SEMAA program were instrumental in shaping my aspirations. Dr. Khandaker's dedication to fostering a passion for STEM children like me inspired me to strive for excellence. His vision and commitment to educational outreach resonated deeply with me and motivated me to pursue a path in computer science.

As a senior majoring in computer science today, I reflect on the pivotal role that the NASA-SEMAA program played in instilling the fundamental values of curiosity, problem-solving, and innovation within me. The exposure to cutting-edge technologies and the emphasis on rigorous STEM curriculum enhancement set the stage for my academic and professional journey.

Jadyn continued his reflections by saying, as I stride toward the horizon of new beginnings, my commitment to innovation and scientific exploration remains. Every accomplishment attained from high school to my senior year as a Computer Science major at the University of Delaware is the resultant of the foundations laid during my time with NASA-SEMAA.

From pioneering initiatives supporting campus communities to contributing to research in advancing solar cell technology, my journey reverberates with the principles instilled in me at a young age. The seeds of curiosity, problem-solving, and a drive for innovation, developed by the immersive STEM exposure of the SEMAA program, continue to guide my trajectory. They form the bedrock upon which I aspire to build a future STEM career with groundbreaking contributions to the world of science and technology.

As I stand at this juncture, I am resolute in my pursuit: to push the boundaries of innovation, unravel the mysteries of science, and contribute meaningfully to the ever-evolving landscape of discovery. Looking ahead to my graduation next semester, I am enthusiastic about continuing to leverage my roles within various campus organizations to further enhance our campus environment. I'm committed to driving positive change and fostering an inclusive atmosphere that supports every individual's growth and success.

Thank you, Jadyn and heartiest congratulations, to you for successfully transitioning from a 4th grade NASA STEM student to passionate and forward-thinking Computer Science major with a strong focus on Artificial Intelligence, Machine Learning and Robotics.

York greatly acknowledges the generous support of Con Edison, AT&T, and National Grid, which enabled both SEMAA and MUREP to offer state-of-the-art STEM education to the greater York community for close to two decades. Con Edison has been a steadfast financial supporter of the previous SEMAA Program since 2007 and continued providing strategic support to the York-hosted STEM Program.

York College enriches lives and enables students to grow as passionate, engaged learners with the confidence to realize their intellectual and human potential as individuals and global citizens. With more than 60 academic majors, York College attracts undergraduate and graduate students from across the five boroughs of New York City as well as Long Island, Westchester County and internationally. More on the NASA-STEM program at York College .

Read more here:

How did the NASA STEM Program trigger curiosity and keep Jadyn Worthington on track? - York College

Gaurav Sukhatme, professor of computer science and electrical and computer engineering, was formally appointed by USC President Carol Folt as the inaugural director of the USC School of Advanced Computing on Wednesday.

Stepping up to the podium to resounding applause from the audience filling the Ronald Tutor Campus Center rotunda, Sukhatme, executive vice dean of the USC Viterbi School of Engineering, said he was lost for words before thanking Folt for her faith in him. I am both honored and touched, he said.

In separate speeches, both Folt and USC Viterbi Dean Yannis C. Yortsos expressed their confidence in Sukhatmes ability to serve as a leader of the universitys 23rd school. For a powerhouse school, we needed a powerhouse director, Folt said. We found that in Gaurav.

The new school, which was officially announced last week, is a keystone of Folts Frontiers of Computing moonshot, serving as a nexus for advanced computing research and education across the university.

In fall 2023, Folt jumpstarted the initiative with a university gift of $260 million from the Lord Foundation of California aimed at advancing and expanding computing research and education across the university.

Our achievements in engineering and computation have already changed the world in so many ways, Folt said. I like to say that USC knows how to move fast when it is needed most. And thats what were seeing and celebrating today.

The new school, a unit of USC Viterbi, will be home to the Thomas Lord Department of Computer Science, the Ming Hsieh Department of Electrical and Computer Engineering, and the Division of Computing Education, but faculty across all schools will be involved in creating this digital backbone, Folt said.

Students are at the heart of the initiative, which Folt said aims to provide knowledge of advanced computing to all, regardless of discipline.

The jobs of the future will require digital skills and computer literacy, she said. Our moonshot, when fully implemented, is determined to ensure that every single student will be prepared to be a part of creating the world of the future.

Describing Sukhatme as a true multihyphenate with a multidisciplinary mind, Folt praised his academic accomplishments and leadership acumen, acknowledging his scholarly works whopping 39,000 citations and his numerous leadership roles at the university.

He also embodies the philosophy of putting students first, she said, highlighting his mentorship of more than 45 doctoral students and 100 masters and undergraduate students during his academic career at USC.

The inauguration also served as an opportunity to celebrate Sukhatmes long and storied Trojan journey.

Sukhatme is a double USC Viterbi alumnus, having earned his masters degree and doctorate in computer science at USC. He was mentored by Professor Emeritus George Bekey, a renowned pioneer in robotic and technology ethics.

Sukhatme became a professor in 2000 and chaired USCs department of computer science from 2012 to 2017 during one of its fastest periods of growth. Since 2017, he has served as the executive vice dean of USC Viterbi.

In 2023, at the Thomas Lord Department of Computer Science naming ceremony, he received the Donald M. Alstadt Chair in Advanced Computing. Some people may think of this as a capstone, but for Gaurav, it was just another milestone that brings us here today, Folt said.

In closing, Folt reflected on the similarities between cricket, Sukhatmes favorite sport, and computer science, his intellectual passion.

They both require strategic planning by people to succeed and enormous patience, said Folt. While things in computing can move quickly like current advances in AI patience is key, especially given the ethical issues at stake.

Before handing the podium to Sukhatme, Folt thanked Yortsos and Geoffrey Garrett, dean of the USC Marshall School of Business. Youve been real partners from the start, she said. It has been such a great opportunity to work with you.

During his address, Sukhatme pledged to move forward USCs rich academic and interdisciplinary legacy with, to use a computer science term, a full-stack approach, building on the universitys long-standing and well-acknowledged strengths across all our schools.

We seek to build a community of scholars dedicated to expanding the frontiers of computing not only in computer science, electrical engineering and other engineering disciplines, but also in the sciences and mathematics, Sukhatme said.

The work of our faculty [at the USC School of Advanced Computing] will profoundly impact virtually every field, from medicine and the health professions to law, business, communications, the arts and humanities.

Speaking to the assembled crowd of trustees, deans, faculty, staff and students, he outlined his unwavering commitment to innovation in computing across all facets of USC, encouraging diverse collaboration and thought, and fostering academic excellence and inclusivity.

I want to emphasize that while the School of Advanced Computing will operate as a unit within the Viterbi School of Engineering, it is a school for everyone, Sukhatme said. It will be home to innovative new programs in computing to ensure that every USC student learns about the explosive modern advances in machine learning and artificial intelligence, while developing a grounding in the fundamentals of computing that underpin these technologies and their applications.

With the opening of the new Dr. Allen and Charlotte Ginsburg Human-Centered Computation Hall this summer, a new home for the School of Advanced Computing, Sukhatme affirmed the emphasis on scholarship with impact at USC.

Our work will be firmly rooted in solving important societal problems that address issues of sustainability, global health, society and governance, and many others, he said.

Sukhatme concluded by offering his deepest gratitude to Yortsos for his remarkable stewardship for nearly two decades.

For his part, Yortsos, during the ceremonys closing speech, said he didnt have to look far to find the perfect leader to entrust with this initiatives extraordinary genesis.

When I was asked who would be the ideal person to run the new school, help it stand, take its first steps and grow to adulthood, I knew that the right person was next to me: Gaurav Sukhatme, he said.

Yortsos said he envisions the new school as a trusted, indivisible and quintessential part of the entire university. A school of advanced computing for all.

Paraphrasing economist W. Brian Arthur, he said: Technology is leveraging phenomena for useful purposes. These words encapsulate all that we do.

The phenomena may vary, he said, from the physical, chemical, geological, biological and planetary, to the behavioral and social.

The useful purposes can address grand challenges and moonshots, but what leverages these phenomena, and the reason we are here today, is advanced computing, Yortsos said. This is, then, the mission of our new school: to leverage phenomena using advanced computing.

That mission, he added, will include the solution of vexing problems in sustainability, health, security; enriching life; and the pruning and ultimate elimination of unavoidable unintended consequences that challenge our ethics and our humanity.

In closing, Yortsos delivered this message to Sukhatme, his friend and colleague of more than 20 years:

I am thrilled to have the opportunity to also usher in this historic moment for the entire university and to congratulate and celebrate our remarkable colleague, who is entrusted with carrying forward a truly remarkable vision.

Read the rest here:

USC celebrates School of Advanced Computing director - University of Southern California