This is a nice tutorial if you are interested in how and why overfitting (etc) happens, how/why regularization helps to alleviate overfitting, ...
"The Theory Behind Overfitting, Cross Validation, Regularization, Bagging, and Boosting: Tutorial", Benyamin Ghojogh, Mark Crowley
One thing I love about #teaching is that you always #learn new things, even when giving a lecture about stuff that you know (or you think you do) very well.
Today a student asked why we should use Student's #t-test at all when comparing samples with equal variances if you can use Welch's variant, which works with unequal variances... that is, why bother checking equality of variances?
Indeed, in the case of equal variances the calculated t statistics would be the same!
This led to a great discussion involving 3 instructors, a bunch of students, and a few simulations in #Rstats which I really enjoyed!
The answer is that Welch's variation also changes how degrees of freedom are calculated, so no they are not equal!
Also, the current recommendation is NOT to check for variance equality (e.g. using Levene's test) as this paper clearly points out, lest you want to increase your Type I error!
https://bpspsychub.onlinelibrary.wiley.com/doi/pdf/10.1348/000711004849222
Gotta go change some slides...
I just discovered this awesome #latex package to #annotate #equations.
https://github.com/st--/annotate-equations/tree/main
Someone is getting some snazzy lectures ready for next week...
I am happy to share our latest #research article, looking at #compassion towards laboratory animals in #undergraduate #biomedical sciences students in the UK and China.
In this study, we developed a survey to explore compassion towards laboratory animals in students and used it in groups of biomedical sciences students in the UK and China.
Exploring Compassion towards Laboratory Animals in UK- and China-Based Undergraduate Biomedical Sciences Students - https://www.mdpi.com/2076-2615/13/22/3584
#bioawk is a command-line gem; it’s an extension of awk that auto-assigns variables for BED, SAM, VCF, GFF, and FASTX[AQ] format files, speeding up routine tasks.
For FASTX:
$1:name
$2:seq
$3:qual (FASTQ only)
$4:comment
Found in @vsbuffalo’s great #Bioinformatics Data Skills.
I am actively recruiting students for my lab at UMass Dartmouth!
Are you interested in a project on actin, amoebae, cell migration, and/or pathogenesis? Send me an email & apply to the Integrative Biology PhD program by Jan 15!
Please RT
More info: https://katrinavelle.wixsite.com/science
What does a #scientist look like? Children are drawing women more than ever before.
@nicolaromano @tomstafford
I'd agree. But that's why I voted C. It's past the time where these self serving bad practices can be called "inadvertent".
Researchers! Please complete my poll with uneven response options (and boost for reach)
How much of a problem is deliberate research fraud?
The problem isn't that COVID is deadly.
The problem is that different people react to it differently.
For most people, COVID is a mild cold. Often, it's so slight that people don't realize they have it.
For some people, it's terrible -- but something that they can handle at home.
For a few people, COVID sends them to the hospital. Or gives them long COVID.
And it killed roughly 1% of people who got it at the beginning of the pandemic, if I recall correctly.
So, if Boris had done this really stupid act, chances are that he would be fine. But many more Britons would have died.
meetings today? zero should be the goal https://blog.codinghorror.com/meetings-where-work-goes-to-die/
"We must not forget that when radium was discovered no one knew that it would prove useful in hospitals. The work was one of pure science. And this is a proof that scientific work must not be considered from the point of view of the direct usefulness of it. It must be done for itself, for the beauty of science, and then there is always the chance that a scientific discovery may become like the radium a benefit for humanity."
Lecture at Vassar College
~Marie Curie (7 November 1867 – 4 July 1934)
Tonight a dear friend sent me this photo of my lab circa 2007. A coffee bar, shelves of books, and off-camera a 12' long table in a massive room surrounded by whiteboards.
Now? My department has moved to a sleek new building. I don't have a lab, just a 120 square foot office. Theorists don't need space; all they do is think. If they need to talk to one another, they can reserve departmental conference rooms so long as they remember to do so six weeks in advance.
Progress marches on.
20 years ago on November 6, 2003, Fedora Linux, originating as "Fedora Core," began in 2003.
Originally referred to as "Fedora Core," Fedora Linux forked from Red Hat Linux in 2003. This move came as Red Hat Linux was phased out, enabling Red Hat to focus on its paid server version, Red Hat Enterprise Linux.
Fedora Linux became a community distribution, while Red Hat Enterprise Linux continued as the official Red Hat-supported distribution.
Last time I brought this up on a UK-based technical mailing list, an entrepreneur—themself from a historically persecuted ethnic group—told me that, "no one cares".
Well, I can't make anyone care but if you are in the position to advise on or choose a domain name you should maybe at least know some of the history behind .IO before buying one.
https://www.beep.blog/io/
"He was doing physical therapy, spending time with family members, and playing cards with his wife, according to the university. But in the days leading up to his death, his heart began to show signs of organ rejection; in other words, his immune system recognized the pig heart as foreign and attacked it. "
https://www.wired.com/story/pig-heart-transplant-lawrence-faucette-death/
Hello #scrnaseq wizards out there! Anyone can help answer this question?
https://www.biostars.org/p/9578398/
I'm happy with the #Seurat + #Signac pipeline but I'm not sure what to do and how to merge the two sets of data (#multiomics and #scRNAseq)
@ct_bergstrom compared to the 999999 other sites that are complete and utterly flaming pieces that contain only a few papers, terrible search and 15 layers of verify your institution before you preview?
The minimal intrinsic stochasticity of constitutively expressed eukaryotic genes is sub-Poissonian
https://www.science.org/doi/full/10.1126/sciadv.adh5138
"Our findings redefine the lower limit of eukaryotic gene expression noise and uncover molecular requirements for achieving ultralow noise, which is expected to be important for vital cellular functions."
Senior lecturer at ZJE and Edinburgh university.
I teach #imageanalysis & #dataanalysis with #RStats & #python. I study #heterogeneity in #pituitary (and other) cells.