I must say it, @K_G_Andersen never ends to surprise me with his compelling arguments... If you want to figure out why Wade is wrong you can Google it 🤣😂🤣

Here he downplays the FCS of SARS2. These fragments do not really come and go all the time. I am not aware of experiments on CoV where the FCS arose by cell or animal passage. More on the FCS here:
https://t.co/oyPKBVVG4f
Here he refers to RmYN02, which has no insertion at the S1/S2 cleavage site
https://t.co/0Goapx6cXj
https://t.co/zEALrUUj8F
And to Gallaher's paper, with the fanciful theory of recombination on a train to Wuhan
Recombination is rare outside its group +
https://t.co/DBs7RMeq4X
As also my co-author @ydeigin says
https://t.co/qrOG0nlIzR
the FCS could be not optimal because part of a vaccine attenuation strategy
https://t.co/tHWG1Igf7Q
Andersen admitted to be wrong with his prediction of the O-linked glycans
The FCS might be similar at the aa level to some FCoV sequences, but less at the nt level.
https://t.co/Hvi3azygu6
Andersen does not mention that the FCS binds extremely well to heparan sulphate as possible result of cell passage
https://t.co/s7q6Ynh7NY
That the "P" is mutating towards residues creating more optimal furin sites could be reversion, which is not rare in live attenuated viruses, mostly if not completely attenuated because accidentally leaked.
"the exact (P)RRAR can be found in other coronaviruses." is true only at the amino acid level, not as nucleotides.
Andersen admits that the codon CGG is rare in CoV because it stimulates an immune reaction. It is only found in 3 % of arginine in SARS2. But he forgets to mention that CGG is double in the FCS, making it very special.
https://t.co/dgrP8nwPRz
Nothing unusual here.
He forgets also to mention that by chance CGG is the best codon for arginine in humans, and CGA, used as second codon in FCoV is not that good (0.11 CGA vs 0.21 CGG)
https://t.co/KQpNwt6QLq
And the fact that the FCS is remarkably stable and necessary for human-to-human transmission in a virus which had only few months to adapt to a new host might hint to previous cell passage in human airways cells, where it is stable or humanized mice.
So all the points from Andersen to disprove Baltimore’s observation on the FCS are false/misleading.
https://t.co/otPnkrF7KS
Actually our @Daoyu15 commented on this, but his replies have been hidden by @Merz. This is the way virologists welcome dissenting opinions.
https://t.co/X8DgaKljk0
https://t.co/THbWA1HhUz
Andersen does not take into account also the other special feature of SARS2, which is its special RBD adapted from the first isolate for very efficient human infection. He tried to justify it with its presence in the pangolin CoV MP789, which we can’t trust.
https://t.co/U2qvG3tXBG
And beside the special FCS and RBD we have a mountain of circumstantial evidence for a lab leak, well explained by Wade, which comprises deception of data and blocking a fair investigation.

More from All

Margatha Natarajar murthi - Uthirakosamangai temple near Ramanathapuram,TN
#ArudraDarisanam
Unique Natarajar made of emerlad is abt 6 feet tall.
It is always covered with sandal paste.Only on Thriuvadhirai Star in month Margazhi-Nataraja can be worshipped without sandal paste.


After removing the sandal paste,day long rituals & various abhishekam will be
https://t.co/e1Ye8DrNWb day Maragatha Nataraja sannandhi will be closed after anointing the murthi with fresh sandal paste.Maragatha Natarajar is covered with sandal paste throughout the year


as Emerald has scientific property of its molecules getting disturbed when exposed to light/water/sound.This is an ancient Shiva temple considered to be 3000 years old -believed to be where Bhagwan Shiva gave Veda gyaana to Parvati Devi.This temple has some stunning sculptures.
How can we use language supervision to learn better visual representations for robotics?

Introducing Voltron: Language-Driven Representation Learning for Robotics!

Paper: https://t.co/gIsRPtSjKz
Models: https://t.co/NOB3cpATYG
Evaluation: https://t.co/aOzQu95J8z

🧵👇(1 / 12)


Videos of humans performing everyday tasks (Something-Something-v2, Ego4D) offer a rich and diverse resource for learning representations for robotic manipulation.

Yet, an underused part of these datasets are the rich, natural language annotations accompanying each video. (2/12)

The Voltron framework offers a simple way to use language supervision to shape representation learning, building off of prior work in representations for robotics like MVP (
https://t.co/Pb0mk9hb4i) and R3M (https://t.co/o2Fkc3fP0e).

The secret is *balance* (3/12)

Starting with a masked autoencoder over frames from these video clips, make a choice:

1) Condition on language and improve our ability to reconstruct the scene.

2) Generate language given the visual representation and improve our ability to describe what's happening. (4/12)

By trading off *conditioning* and *generation* we show that we can learn 1) better representations than prior methods, and 2) explicitly shape the balance of low and high-level features captured.

Why is the ability to shape this balance important? (5/12)

You May Also Like